Menu
▾
▴
host-sflow-discuss — host-sflow discussion list
You can subscribe to this list here.
| 2010 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
(12) |
Jul
|
Aug
|
Sep
(4) |
Oct
(5) |
Nov
|
Dec
(5) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 |
Jan
(34) |
Feb
(67) |
Mar
(27) |
Apr
|
May
(13) |
Jun
(9) |
Jul
(14) |
Aug
(4) |
Sep
(6) |
Oct
(3) |
Nov
|
Dec
|
| 2012 |
Jan
(7) |
Feb
|
Mar
(29) |
Apr
(28) |
May
(3) |
Jun
(8) |
Jul
(36) |
Aug
(64) |
Sep
|
Oct
(3) |
Nov
(2) |
Dec
(14) |
| 2013 |
Jan
(12) |
Feb
(6) |
Mar
|
Apr
(5) |
May
(2) |
Jun
(2) |
Jul
(10) |
Aug
(2) |
Sep
(7) |
Oct
(5) |
Nov
(2) |
Dec
(3) |
| 2014 |
Jan
(4) |
Feb
(5) |
Mar
|
Apr
(1) |
May
|
Jun
(2) |
Jul
|
Aug
|
Sep
(4) |
Oct
(6) |
Nov
(4) |
Dec
(4) |
| 2015 |
Jan
|
Feb
|
Mar
(3) |
Apr
(6) |
May
(8) |
Jun
(5) |
Jul
(16) |
Aug
|
Sep
(1) |
Oct
|
Nov
(4) |
Dec
|
| 2016 |
Jan
(3) |
Feb
|
Mar
(5) |
Apr
|
May
|
Jun
(3) |
Jul
|
Aug
(1) |
Sep
(4) |
Oct
|
Nov
|
Dec
|
|
From: Tony V. (PCNET) <to...@pc...> - 2016-09-19 18:33:08
|
Unsubscribe please -----Original Message----- From: hos...@li... [mailto:hos...@li...] Sent: Monday, September 19, 2016 10:02 AM To: hos...@li... Subject: host-sflow-discuss Digest, Vol 48, Issue 1 Send host-sflow-discuss mailing list submissions to hos...@li... To subscribe or unsubscribe via the World Wide Web, visit https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss or, via email, send a message with subject or body 'help' to hos...@li... You can reach the person managing the list at hos...@li... When replying, please edit your Subject line so it is more specific than "Re: Contents of host-sflow-discuss digest..." Today's Topics: 1. FW: Oracle Database (Bei, Richard E) 2. PCAP in promiscous mode (Max Grobecker) 3. Re: PCAP in promiscous mode (Neil McKee) 4. Re: PCAP in promiscous mode (Max Grobecker) ---------------------------------------------------------------------- Message: 1 Date: Thu, 4 Aug 2016 16:24:21 +0000 From: "Bei, Richard E" <ric...@lm...> Subject: [host-sflow-discuss] FW: Oracle Database To: "'hos...@li...'" <hos...@li...> Message-ID: <230...@HV...> Content-Type: text/plain; charset="us-ascii" From: Bei, Richard E (US N-ETG) Sent: Thursday, August 04, 2016 12:20 PM To: 'hos...@li....' <hos...@li....> Subject: Oracle Database I would like to know if the metrics that are created can be exported into an oracle database. VR; Richard Bei Systems Administrator MCSE/ MCSA/ MCP/ SEC + Lockheed Martin Corporation Ric...@lm...<mailto:Ric...@lm...> (O) 703.466.2796 -------------- next part -------------- An HTML attachment was scrubbed... ------------------------------ Message: 2 Date: Sun, 18 Sep 2016 18:02:11 +0200 From: Max Grobecker <max...@ml...> Subject: [host-sflow-discuss] PCAP in promiscous mode To: hos...@li... Message-ID: <78b...@ml...> Content-Type: text/plain; charset="iso-8859-15" Hello, I have a test setup on a KVM host with bridged interfaces, installed version is hsflowd version 2.0.4. eth0 is the physical device, breth0 is the bridge whith eth0 and the KVM network interfaces attached to it. breth0 has an IP address assigned to reach the hosts SSH console. Now I want to create flows of the whole traffic that passes the eth0 interface, so my config looks like: ---------------- sflow { agent = breth0 polling = 10 sampling = 400 pcap { dev = eth0 } } ---------------- I've tested also with "dev = breth0" and "dev = vm26" (a virtual machine's device), but nothing helped. When I start hsflowd with debugging I can see that packets flowing through the eth0 interface are seen, but they seem not to be recognized and are not exported as a flow. All I can see in the flows is SSH traffic of the host... Only packets with a MAC address directly assigned to the KVM host are exported as a flow, not the packets that fly through the bridge to one of the virtual machines. Is there anything I can do to also export these packets? That's what I can see on the console when running in debug mode: ---------------- takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=000A8A974580, macdst=5254029E4544 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, macdst=5254029E4544 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, macdst=52540275CE51 takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 cap_len=52 mac_len=14 000A8A974580 -> 52540275CE51 (ethertype=0x0800) macsrc=000A8A974580, macdst=5254029E4544 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, macdst=5254029E4544 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=5254029E4544, macdst=000A8A974580 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=5254029E4544, macdst=000A8A974580 takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=52540275CE51, macdst=000A8A974580 ---------------- Thanks for your help! Max -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature ------------------------------ Message: 3 Date: Sun, 18 Sep 2016 11:02:56 -0700 From: Neil McKee <nei...@in...> Subject: Re: [host-sflow-discuss] PCAP in promiscous mode To: Max Grobecker <max...@ml...> Cc: "hos...@li..." <hos...@li...> Message-ID: <CAB...@ma...> Content-Type: text/plain; charset="utf-8" You can change: pcap { dev = eth0 } to: pcap { dev = eth0 promisc = on } but it's really much better to run Open vSwitch as the KVM virtual switch and then replace the pcap {} section in /etc/hsflowd.conf with this: ovs { } That enables mod_ovs in hsflowd, which simply shares the sFlow configuration with Open vSwitch automatically whenever it changes. Now hsflowd will report the hypervisor + vm-stats while OVS reports packet-samples and vport counters for all its ports. (Open vSwitch has included sFlow since 2009). Neil On Sun, Sep 18, 2016 at 9:02 AM, Max Grobecker < max...@ml...> wrote: > Hello, > > I have a test setup on a KVM host with bridged interfaces, installed > version is hsflowd version 2.0.4. > eth0 is the physical device, breth0 is the bridge whith eth0 and the > KVM network interfaces attached to it. > breth0 has an IP address assigned to reach the hosts SSH console. > > Now I want to create flows of the whole traffic that passes the eth0 > interface, so my config looks like: > > ---------------- > sflow { > agent = breth0 > > polling = 10 > sampling = 400 > > pcap { dev = eth0 } > } > ---------------- > > I've tested also with "dev = breth0" and "dev = vm26" (a virtual > machine's device), but nothing helped. > > When I start hsflowd with debugging I can see that packets flowing > through the eth0 interface are seen, but they seem not to be > recognized and are not exported as a flow. > All I can see in the flows is SSH traffic of the host... > > Only packets with a MAC address directly assigned to the KVM host are > exported as a flow, not the packets that fly through the bridge to one > of the virtual machines. > Is there anything I can do to also export these packets? > > That's what I can see on the console when running in debug mode: > > ---------------- > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=000A8A974580, > macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, > macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, > macdst=52540275CE51 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 > cap_len=52 > mac_len=14 > 000A8A974580 -> 52540275CE51 (ethertype=0x0800) macsrc=000A8A974580, > macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=000A8A974580, > macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) macsrc=5254029E4544, > macdst=000A8A974580 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=5254029E4544, > macdst=000A8A974580 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 > cap_len=76 > mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) macsrc=52540275CE51, > macdst=000A8A974580 > ---------------- > > > > Thanks for your help! > Max > > > ------------------------------------------------------------ > ------------------ > > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... ------------------------------ Message: 4 Date: Mon, 19 Sep 2016 20:02:18 +0200 From: Max Grobecker <max...@ml...> Subject: Re: [host-sflow-discuss] PCAP in promiscous mode To: Neil McKee <nei...@in...> Cc: "hos...@li..." <hos...@li...> Message-ID: <4d2...@ml...> Content-Type: text/plain; charset="utf-8" Hi Neil, thanks - that did the trick :-) I'll have a look at Open vSwitch. Greetings Max Am 18.09.2016 um 20:02 schrieb Neil McKee: > You can change: > > pcap { dev = eth0 } > > to: > > pcap { > dev = eth0 > promisc = on > } > > but it's really much better to run Open vSwitch as the KVM virtual switch and then replace the pcap {} section in /etc/hsflowd.conf with this: > > ovs { } > > That enables mod_ovs in hsflowd, which simply shares the sFlow configuration with Open vSwitch automatically whenever it changes. Now hsflowd will report the hypervisor + vm-stats while OVS reports packet-samples and vport counters for all its ports. (Open vSwitch has included sFlow since 2009). > > Neil > > > On Sun, Sep 18, 2016 at 9:02 AM, Max Grobecker <max...@ml... <mailto:max...@ml...>> wrote: > > Hello, > > I have a test setup on a KVM host with bridged interfaces, installed version is hsflowd version 2.0.4. > eth0 is the physical device, breth0 is the bridge whith eth0 and the KVM network interfaces attached to it. > breth0 has an IP address assigned to reach the hosts SSH console. > > Now I want to create flows of the whole traffic that passes the eth0 interface, so my config looks like: > > ---------------- > sflow { > agent = breth0 > > polling = 10 > sampling = 400 > > pcap { dev = eth0 } > } > ---------------- > > I've tested also with "dev = breth0" and "dev = vm26" (a virtual machine's device), but nothing helped. > > When I start hsflowd with debugging I can see that packets flowing through the eth0 interface are seen, > but they seem not to be recognized and are not exported as a flow. > All I can see in the flows is SSH traffic of the host... > > Only packets with a MAC address directly assigned to the KVM host are exported as a flow, not the packets > that fly through the bridge to one of the virtual machines. > Is there anything I can do to also export these packets? > > That's what I can see on the console when running in debug mode: > > ---------------- > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) > macsrc=000A8A974580, macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) > macsrc=000A8A974580, macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) > macsrc=000A8A974580, macdst=52540275CE51 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 cap_len=52 mac_len=14 > 000A8A974580 -> 52540275CE51 (ethertype=0x0800) > macsrc=000A8A974580, macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) > macsrc=000A8A974580, macdst=5254029E4544 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 000A8A974580 -> 5254029E4544 (ethertype=0x0800) > macsrc=5254029E4544, macdst=000A8A974580 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) > macsrc=5254029E4544, macdst=000A8A974580 > takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14 > 5254029E4544 -> 000A8A974580 (ethertype=0x0800) > macsrc=52540275CE51, macdst=000A8A974580 > ---------------- > > > > Thanks for your help! > Max > > > > ---------------------------------------------------------------------- > -------- > > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... <mailto:hos...@li...> > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > <https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss> > > -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature ------------------------------ ------------------------------------------------------------------------------ ------------------------------ _______________________________________________ host-sflow-discuss mailing list hos...@li... https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss End of host-sflow-discuss Digest, Vol 48, Issue 1 ************************************************* |
|
From: Max G. <max...@ml...> - 2016-09-19 18:02:27
|
Hi Neil,
thanks - that did the trick :-)
I'll have a look at Open vSwitch.
Greetings
Max
Am 18.09.2016 um 20:02 schrieb Neil McKee:
> You can change:
>
> pcap { dev = eth0 }
>
> to:
>
> pcap {
> dev = eth0
> promisc = on
> }
>
> but it's really much better to run Open vSwitch as the KVM virtual switch and then replace the pcap {} section in /etc/hsflowd.conf with this:
>
> ovs { }
>
> That enables mod_ovs in hsflowd, which simply shares the sFlow configuration with Open vSwitch automatically whenever it changes. Now hsflowd will report the hypervisor + vm-stats while OVS reports packet-samples and vport counters for all its ports. (Open vSwitch has included sFlow since 2009).
>
> Neil
>
>
> On Sun, Sep 18, 2016 at 9:02 AM, Max Grobecker <max...@ml... <mailto:max...@ml...>> wrote:
>
> Hello,
>
> I have a test setup on a KVM host with bridged interfaces, installed version is hsflowd version 2.0.4.
> eth0 is the physical device, breth0 is the bridge whith eth0 and the KVM network interfaces attached to it.
> breth0 has an IP address assigned to reach the hosts SSH console.
>
> Now I want to create flows of the whole traffic that passes the eth0 interface, so my config looks like:
>
> ----------------
> sflow {
> agent = breth0
>
> polling = 10
> sampling = 400
>
> pcap { dev = eth0 }
> }
> ----------------
>
> I've tested also with "dev = breth0" and "dev = vm26" (a virtual machine's device), but nothing helped.
>
> When I start hsflowd with debugging I can see that packets flowing through the eth0 interface are seen,
> but they seem not to be recognized and are not exported as a flow.
> All I can see in the flows is SSH traffic of the host...
>
> Only packets with a MAC address directly assigned to the KVM host are exported as a flow, not the packets
> that fly through the bridge to one of the virtual machines.
> Is there anything I can do to also export these packets?
>
> That's what I can see on the console when running in debug mode:
>
> ----------------
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=52540275CE51
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 cap_len=52 mac_len=14
> 000A8A974580 -> 52540275CE51 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=5254029E4544, macdst=000A8A974580
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=5254029E4544, macdst=000A8A974580
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=52540275CE51, macdst=000A8A974580
> ----------------
>
>
>
> Thanks for your help!
> Max
>
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> host-sflow-discuss mailing list
> hos...@li... <mailto:hos...@li...>
> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss <https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss>
>
>
|
|
From: Neil M. <nei...@in...> - 2016-09-18 18:30:33
|
You can change:
pcap { dev = eth0 }
to:
pcap {
dev = eth0
promisc = on
}
but it's really much better to run Open vSwitch as the KVM virtual switch
and then replace the pcap {} section in /etc/hsflowd.conf with this:
ovs { }
That enables mod_ovs in hsflowd, which simply shares the sFlow
configuration with Open vSwitch automatically whenever it changes. Now
hsflowd will report the hypervisor + vm-stats while OVS reports
packet-samples and vport counters for all its ports. (Open vSwitch has
included sFlow since 2009).
Neil
On Sun, Sep 18, 2016 at 9:02 AM, Max Grobecker <
max...@ml...> wrote:
> Hello,
>
> I have a test setup on a KVM host with bridged interfaces, installed
> version is hsflowd version 2.0.4.
> eth0 is the physical device, breth0 is the bridge whith eth0 and the KVM
> network interfaces attached to it.
> breth0 has an IP address assigned to reach the hosts SSH console.
>
> Now I want to create flows of the whole traffic that passes the eth0
> interface, so my config looks like:
>
> ----------------
> sflow {
> agent = breth0
>
> polling = 10
> sampling = 400
>
> pcap { dev = eth0 }
> }
> ----------------
>
> I've tested also with "dev = breth0" and "dev = vm26" (a virtual machine's
> device), but nothing helped.
>
> When I start hsflowd with debugging I can see that packets flowing through
> the eth0 interface are seen,
> but they seem not to be recognized and are not exported as a flow.
> All I can see in the flows is SSH traffic of the host...
>
> Only packets with a MAC address directly assigned to the KVM host are
> exported as a flow, not the packets
> that fly through the bridge to one of the virtual machines.
> Is there anything I can do to also export these packets?
>
> That's what I can see on the console when running in debug mode:
>
> ----------------
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=52540275CE51
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 cap_len=52
> mac_len=14
> 000A8A974580 -> 52540275CE51 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=000A8A974580, macdst=5254029E4544
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 000A8A974580 -> 5254029E4544 (ethertype=0x0800)
> macsrc=5254029E4544, macdst=000A8A974580
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=5254029E4544, macdst=000A8A974580
> takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76
> mac_len=14
> 5254029E4544 -> 000A8A974580 (ethertype=0x0800)
> macsrc=52540275CE51, macdst=000A8A974580
> ----------------
>
>
>
> Thanks for your help!
> Max
>
>
> ------------------------------------------------------------
> ------------------
>
> _______________________________________________
> host-sflow-discuss mailing list
> hos...@li...
> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss
>
>
|
|
From: Max G. <max...@ml...> - 2016-09-18 16:19:04
|
Hello,
I have a test setup on a KVM host with bridged interfaces, installed version is hsflowd version 2.0.4.
eth0 is the physical device, breth0 is the bridge whith eth0 and the KVM network interfaces attached to it.
breth0 has an IP address assigned to reach the hosts SSH console.
Now I want to create flows of the whole traffic that passes the eth0 interface, so my config looks like:
----------------
sflow {
agent = breth0
polling = 10
sampling = 400
pcap { dev = eth0 }
}
----------------
I've tested also with "dev = breth0" and "dev = vm26" (a virtual machine's device), but nothing helped.
When I start hsflowd with debugging I can see that packets flowing through the eth0 interface are seen,
but they seem not to be recognized and are not exported as a flow.
All I can see in the flows is SSH traffic of the host...
Only packets with a MAC address directly assigned to the KVM host are exported as a flow, not the packets
that fly through the bridge to one of the virtual machines.
Is there anything I can do to also export these packets?
That's what I can see on the console when running in debug mode:
----------------
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
5254029E4544 -> 000A8A974580 (ethertype=0x0800)
macsrc=000A8A974580, macdst=5254029E4544
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
000A8A974580 -> 5254029E4544 (ethertype=0x0800)
macsrc=000A8A974580, macdst=5254029E4544
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
000A8A974580 -> 5254029E4544 (ethertype=0x0800)
macsrc=000A8A974580, macdst=52540275CE51
takeSample: hook=0 in=<not found> out=<not found> pkt_len=66 cap_len=52 mac_len=14
000A8A974580 -> 52540275CE51 (ethertype=0x0800)
macsrc=000A8A974580, macdst=5254029E4544
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
000A8A974580 -> 5254029E4544 (ethertype=0x0800)
macsrc=000A8A974580, macdst=5254029E4544
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
000A8A974580 -> 5254029E4544 (ethertype=0x0800)
macsrc=5254029E4544, macdst=000A8A974580
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
5254029E4544 -> 000A8A974580 (ethertype=0x0800)
macsrc=5254029E4544, macdst=000A8A974580
takeSample: hook=0 in=<not found> out=<not found> pkt_len=90 cap_len=76 mac_len=14
5254029E4544 -> 000A8A974580 (ethertype=0x0800)
macsrc=52540275CE51, macdst=000A8A974580
----------------
Thanks for your help!
Max
|
|
From: Bei, R. E <ric...@lm...> - 2016-08-04 16:24:50
|
From: Bei, Richard E (US N-ETG) Sent: Thursday, August 04, 2016 12:20 PM To: 'hos...@li....' <hos...@li....> Subject: Oracle Database I would like to know if the metrics that are created can be exported into an oracle database. VR; Richard Bei Systems Administrator MCSE/ MCSA/ MCP/ SEC + Lockheed Martin Corporation Ric...@lm...<mailto:Ric...@lm...> (O) 703.466.2796 |
|
From: Brian J. <bri...@vt...> - 2016-06-10 15:38:14
|
I will give that a try. Thanks -- Brian On Thu, Jun 9, 2016 at 10:38 AM, Neil McKee <nei...@in...> wrote: > There is a partial implementation in the source tree under src/Darwin. If > you build that you will get the executable, but as yet no one has added > the .dmg packaging or launcher file. If you know how to do that stuff, > please go ahead! > > Neil > > > On Thursday, June 9, 2016, Brian Jones <bri...@vt...> wrote: > >> Any thoughts of getting host sFlow going on the MAC? >> >> Thanks, >> >> -- >> Brian >> > > > -- > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > |
|
From: Neil M. <nei...@in...> - 2016-06-09 15:01:42
|
There is a partial implementation in the source tree under src/Darwin. If you build that you will get the executable, but as yet no one has added the .dmg packaging or launcher file. If you know how to do that stuff, please go ahead! Neil On Thursday, June 9, 2016, Brian Jones <bri...@vt...> wrote: > Any thoughts of getting host sFlow going on the MAC? > > Thanks, > > -- > Brian > -- ------ Neil McKee InMon Corp. http://www.inmon.com |
|
From: Brian J. <bri...@vt...> - 2016-06-09 14:18:41
|
Any thoughts of getting host sFlow going on the MAC? Thanks, -- Brian |
|
From: Tony V. (PCNET) <to...@pc...> - 2016-03-09 07:23:41
|
Remove / unsubscribe Thx Sent from my iPhone > On Mar 8, 2016, at 9:28 PM, "hos...@li..." <hos...@li...> wrote: > > Send host-sflow-discuss mailing list submissions to > hos...@li... > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > or, via email, send a message with subject or body 'help' to > hos...@li... > > You can reach the person managing the list at > hos...@li... > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of host-sflow-discuss digest..." > > > Today's Topics: > > 1. headerBytes configuration in host-sflow (Viswanathan, Vasanth) > 2. Re: headerBytes configuration in host-sflow (Neil McKee) > 3. Re: headerBytes configuration in host-sflow (Viswanathan, Vasanth) > 4. Re: headerBytes configuration in host-sflow (Neil McKee) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 8 Mar 2016 18:25:24 +0000 > From: "Viswanathan, Vasanth" <vas...@hp...> > Subject: [host-sflow-discuss] headerBytes configuration in host-sflow > To: "hos...@li..." > <hos...@li...> > Message-ID: > <6E9...@G4...> > > Content-Type: text/plain; charset="us-ascii" > > Hi everyone, > > I posted a question on the sFlow google group and was redirected to this mailing list to get a better answer. > I would appreciate any kind of help/clarification. > > 1. headerBytes configuration > I am trying to configure host-sflow to use 'headerBytes' parameter. > I am writing the configuration to /etc/hsflowd.conf but the packets I receive at the collector (sflowtool) do not use the configured headerBytes. > For ex: headerBytes = 256, This configuration does not seem to take effect and I only see the default header bytes of 128 from sampled packets. > Doing a grep on the source code shows that the 'headerBytes' parameter is read from the config file but does not seem to be used (Please correct me if I missed it) > > Is the functionality already supported? > If not, any timelines on when we can expect it? > > 2. datagramBytes configuration > I also plan to use 'datagramBytes' parameter to control the max datagram size. > Is this already supported? > > Thanks for your time! > Vasanth. > > > > ------------------------------ > > Message: 2 > Date: Tue, 8 Mar 2016 17:07:50 -0800 > From: Neil McKee <nei...@in...> > Subject: Re: [host-sflow-discuss] headerBytes configuration in > host-sflow > To: "Viswanathan, Vasanth" <vas...@hp...> > Cc: "hos...@li..." > <hos...@li...> > Message-ID: > <CAB...@ma...> > Content-Type: text/plain; charset="utf-8" > > I checked in changes to honor both of these settings -- within the > compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. > > I haven't tested all variations, so please let me know if you find any > problems. > > Neil > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth < > vas...@hp...> wrote: > >> Hi everyone, >> >> I posted a question on the sFlow google group and was redirected to this >> mailing list to get a better answer. >> I would appreciate any kind of help/clarification. >> >> 1. headerBytes configuration >> I am trying to configure host-sflow to use 'headerBytes' parameter. >> I am writing the configuration to /etc/hsflowd.conf but the packets I >> receive at the collector (sflowtool) do not use the configured headerBytes. >> For ex: headerBytes = 256, This configuration does not seem to take effect >> and I only see the default header bytes of 128 from sampled packets. >> Doing a grep on the source code shows that the 'headerBytes' parameter is >> read from the config file but does not seem to be used (Please correct me >> if I missed it) >> >> Is the functionality already supported? >> If not, any timelines on when we can expect it? >> >> 2. datagramBytes configuration >> I also plan to use 'datagramBytes' parameter to control the max datagram >> size. >> Is this already supported? >> >> Thanks for your time! >> Vasanth. >> >> >> ------------------------------------------------------------------------------ >> Transform Data into Opportunity. >> Accelerate data analysis in your applications with >> Intel Data Analytics Acceleration Library. >> Click to learn more. >> http://makebettercode.com/inteldaal-eval >> _______________________________________________ >> host-sflow-discuss mailing list >> hos...@li... >> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > -------------- next part -------------- > An HTML attachment was scrubbed... > > ------------------------------ > > Message: 3 > Date: Wed, 9 Mar 2016 02:45:28 +0000 > From: "Viswanathan, Vasanth" <vas...@hp...> > Subject: Re: [host-sflow-discuss] headerBytes configuration in > host-sflow > To: Neil McKee <nei...@in...> > Cc: "hos...@li..." > <hos...@li...> > Message-ID: > <6E9...@G4...> > > Content-Type: text/plain; charset="utf-8" > > Hi Neil, > > The 'datagramBytes' parameter is working as expected. > I am on Linux platform. But I still don?t see 'headerBytes' taking effect. > I did a grep on host-sflow code. > > $ grep -ir headerBytes src/Linux/* > samplingRate:src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == NULL) return NO; > src/Linux/hsflowd.h: uint32_t headerBytes; > src/Linux/hsflowtokens.h:HSPTOKEN_DATA( HSPTOKEN_HEADERBYTES, "headerBytes", HSPTOKENTYPE_ATTRIB ) > src/Linux/hsflowconfig.c: st->headerBytes = SFL_DEFAULT_HEADER_SIZE; > src/Linux/hsflowconfig.c: case HSPTOKEN_HEADERBYTES: > src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == NULL) return NO; > > With some further digging around, > sflow_sampler.c -> sfl_sampler_set_sFlowFsMaximumHeaderSize(sampler, SFL_DEFAULT_HEADER_SIZE); > readPackets.c -> uint32_t maxHdrLen = sampler->sFlowFsMaximumHeaderSize; > > It seems to read the 'headerBytes' data but not do anything with it. > > I also found that in Windows platform code, the parameter is (maybe?) used. > samplingRate:src/Windows/hsflowd/hypervSwitch.c: sFlowConfig.sampleHeaderLength = sp->sFlow->sFlowSettings->headerBytes; > > Thanks, > Vasanth. > > From: Neil McKee [mailto:nei...@in...] > Sent: Tuesday, March 08, 2016 5:08 PM > To: Viswanathan, Vasanth > Cc: hos...@li... > Subject: Re: [host-sflow-discuss] headerBytes configuration in host-sflow > > I checked in changes to honor both of these settings -- within the compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. > > I haven't tested all variations, ?so please let me know if you find any problems. > > Neil > > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth <vas...@hp...> wrote: > Hi everyone, > > I posted a question on the sFlow google group and was redirected to this mailing list to get a better answer. > I would appreciate any kind of help/clarification. > > 1. headerBytes configuration > I am trying to configure host-sflow to use 'headerBytes' parameter. > I am writing the configuration to /etc/hsflowd.conf but the packets I receive at the collector (sflowtool) do not use the configured headerBytes. > For ex: headerBytes = 256, This configuration does not seem to take effect and I only see the default header bytes of 128 from sampled packets. > Doing a grep on the source code shows that the 'headerBytes' parameter is read from the config file but does not seem to be used (Please correct me if I missed it) > > Is the functionality already supported? > If not, any timelines on when we can expect it? > > 2. datagramBytes configuration > I also plan to use 'datagramBytes' parameter to control the max datagram size. > Is this already supported? > > Thanks for your time! > Vasanth. > > ------------------------------------------------------------------------------ > Transform Data into Opportunity. > Accelerate data analysis in your applications with > Intel Data Analytics Acceleration Library. > Click to learn more. > http://makebettercode.com/inteldaal-eval > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > > > ------------------------------ > > Message: 4 > Date: Tue, 8 Mar 2016 22:28:04 -0800 > From: Neil McKee <nei...@in...> > Subject: Re: [host-sflow-discuss] headerBytes configuration in > host-sflow > To: "Viswanathan, Vasanth" <vas...@hp...> > Cc: "hos...@li..." > <hos...@li...> > Message-ID: > <CAB...@ma...> > Content-Type: text/plain; charset="utf-8" > > Sorry I wasn't clear: I only checked these changes into the master branch > of the source code. I didn't build a new release. So please run: > > git clone https://github.com/sflow/host-sflow > cd host-sflow > make PCAP=yes > make install > > and then try it again. > > You'll find that headerBytes is still limited to being <= 256 by > the HSP_MAX_HEADER_BYTES setting in hsflowd.h, but if you try adding > "headerBytes=256" to /etc/hsflowd.conf and generate some traffic then you > should see packet samples where more than the default limit of 128 bytes > has been captured. If you don't see that, then please send me the > details of your setup. > > Neil > > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Tue, Mar 8, 2016 at 6:45 PM, Viswanathan, Vasanth < > vas...@hp...> wrote: > >> Hi Neil, >> >> The 'datagramBytes' parameter is working as expected. >> I am on Linux platform. But I still don?t see 'headerBytes' taking effect. >> I did a grep on host-sflow code. >> >> $ grep -ir headerBytes src/Linux/* >> samplingRate:src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, >> tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) >> == NULL) return NO; >> src/Linux/hsflowd.h: uint32_t headerBytes; >> src/Linux/hsflowtokens.h:HSPTOKEN_DATA( HSPTOKEN_HEADERBYTES, >> "headerBytes", HSPTOKENTYPE_ATTRIB ) >> src/Linux/hsflowconfig.c: st->headerBytes = SFL_DEFAULT_HEADER_SIZE; >> src/Linux/hsflowconfig.c: case HSPTOKEN_HEADERBYTES: >> src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, >> &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == >> NULL) return NO; >> >> With some further digging around, >> sflow_sampler.c -> sfl_sampler_set_sFlowFsMaximumHeaderSize(sampler, >> SFL_DEFAULT_HEADER_SIZE); >> readPackets.c -> uint32_t maxHdrLen = sampler->sFlowFsMaximumHeaderSize; >> >> It seems to read the 'headerBytes' data but not do anything with it. >> >> I also found that in Windows platform code, the parameter is (maybe?) used. >> samplingRate:src/Windows/hsflowd/hypervSwitch.c: >> sFlowConfig.sampleHeaderLength = sp->sFlow->sFlowSettings->headerBytes; >> >> Thanks, >> Vasanth. >> >> From: Neil McKee [mailto:nei...@in...] >> Sent: Tuesday, March 08, 2016 5:08 PM >> To: Viswanathan, Vasanth >> Cc: hos...@li... >> Subject: Re: [host-sflow-discuss] headerBytes configuration in host-sflow >> >> I checked in changes to honor both of these settings -- within the >> compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. >> >> I haven't tested all variations, so please let me know if you find any >> problems. >> >> Neil >> >> >> >> ------ >> Neil McKee >> InMon Corp. >> http://www.inmon.com >> >> On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth < >> vas...@hp...> wrote: >> Hi everyone, >> >> I posted a question on the sFlow google group and was redirected to this >> mailing list to get a better answer. >> I would appreciate any kind of help/clarification. >> >> 1. headerBytes configuration >> I am trying to configure host-sflow to use 'headerBytes' parameter. >> I am writing the configuration to /etc/hsflowd.conf but the packets I >> receive at the collector (sflowtool) do not use the configured headerBytes. >> For ex: headerBytes = 256, This configuration does not seem to take effect >> and I only see the default header bytes of 128 from sampled packets. >> Doing a grep on the source code shows that the 'headerBytes' parameter is >> read from the config file but does not seem to be used (Please correct me >> if I missed it) >> >> Is the functionality already supported? >> If not, any timelines on when we can expect it? >> >> 2. datagramBytes configuration >> I also plan to use 'datagramBytes' parameter to control the max datagram >> size. >> Is this already supported? >> >> Thanks for your time! >> Vasanth. >> >> >> ------------------------------------------------------------------------------ >> Transform Data into Opportunity. >> Accelerate data analysis in your applications with >> Intel Data Analytics Acceleration Library. >> Click to learn more. >> http://makebettercode.com/inteldaal-eval >> _______________________________________________ >> host-sflow-discuss mailing list >> hos...@li... >> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > -------------- next part -------------- > An HTML attachment was scrubbed... > > ------------------------------ > > ------------------------------------------------------------------------------ > Transform Data into Opportunity. > Accelerate data analysis in your applications with > Intel Data Analytics Acceleration Library. > Click to learn more. > http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140 > > ------------------------------ > > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > > > End of host-sflow-discuss Digest, Vol 46, Issue 1 > ************************************************* |
|
From: Neil M. <nei...@in...> - 2016-03-09 06:28:31
|
Sorry I wasn't clear: I only checked these changes into the master branch of the source code. I didn't build a new release. So please run: git clone https://github.com/sflow/host-sflow cd host-sflow make PCAP=yes make install and then try it again. You'll find that headerBytes is still limited to being <= 256 by the HSP_MAX_HEADER_BYTES setting in hsflowd.h, but if you try adding "headerBytes=256" to /etc/hsflowd.conf and generate some traffic then you should see packet samples where more than the default limit of 128 bytes has been captured. If you don't see that, then please send me the details of your setup. Neil ------ Neil McKee InMon Corp. http://www.inmon.com On Tue, Mar 8, 2016 at 6:45 PM, Viswanathan, Vasanth < vas...@hp...> wrote: > Hi Neil, > > The 'datagramBytes' parameter is working as expected. > I am on Linux platform. But I still don’t see 'headerBytes' taking effect. > I did a grep on host-sflow code. > > $ grep -ir headerBytes src/Linux/* > samplingRate:src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, > tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) > == NULL) return NO; > src/Linux/hsflowd.h: uint32_t headerBytes; > src/Linux/hsflowtokens.h:HSPTOKEN_DATA( HSPTOKEN_HEADERBYTES, > "headerBytes", HSPTOKENTYPE_ATTRIB ) > src/Linux/hsflowconfig.c: st->headerBytes = SFL_DEFAULT_HEADER_SIZE; > src/Linux/hsflowconfig.c: case HSPTOKEN_HEADERBYTES: > src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, > &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == > NULL) return NO; > > With some further digging around, > sflow_sampler.c -> sfl_sampler_set_sFlowFsMaximumHeaderSize(sampler, > SFL_DEFAULT_HEADER_SIZE); > readPackets.c -> uint32_t maxHdrLen = sampler->sFlowFsMaximumHeaderSize; > > It seems to read the 'headerBytes' data but not do anything with it. > > I also found that in Windows platform code, the parameter is (maybe?) used. > samplingRate:src/Windows/hsflowd/hypervSwitch.c: > sFlowConfig.sampleHeaderLength = sp->sFlow->sFlowSettings->headerBytes; > > Thanks, > Vasanth. > > From: Neil McKee [mailto:nei...@in...] > Sent: Tuesday, March 08, 2016 5:08 PM > To: Viswanathan, Vasanth > Cc: hos...@li... > Subject: Re: [host-sflow-discuss] headerBytes configuration in host-sflow > > I checked in changes to honor both of these settings -- within the > compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. > > I haven't tested all variations, so please let me know if you find any > problems. > > Neil > > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth < > vas...@hp...> wrote: > Hi everyone, > > I posted a question on the sFlow google group and was redirected to this > mailing list to get a better answer. > I would appreciate any kind of help/clarification. > > 1. headerBytes configuration > I am trying to configure host-sflow to use 'headerBytes' parameter. > I am writing the configuration to /etc/hsflowd.conf but the packets I > receive at the collector (sflowtool) do not use the configured headerBytes. > For ex: headerBytes = 256, This configuration does not seem to take effect > and I only see the default header bytes of 128 from sampled packets. > Doing a grep on the source code shows that the 'headerBytes' parameter is > read from the config file but does not seem to be used (Please correct me > if I missed it) > > Is the functionality already supported? > If not, any timelines on when we can expect it? > > 2. datagramBytes configuration > I also plan to use 'datagramBytes' parameter to control the max datagram > size. > Is this already supported? > > Thanks for your time! > Vasanth. > > > ------------------------------------------------------------------------------ > Transform Data into Opportunity. > Accelerate data analysis in your applications with > Intel Data Analytics Acceleration Library. > Click to learn more. > http://makebettercode.com/inteldaal-eval > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > > |
|
From: Viswanathan, V. <vas...@hp...> - 2016-03-09 02:45:39
|
Hi Neil, The 'datagramBytes' parameter is working as expected. I am on Linux platform. But I still don’t see 'headerBytes' taking effect. I did a grep on host-sflow code. $ grep -ir headerBytes src/Linux/* samplingRate:src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == NULL) return NO; src/Linux/hsflowd.h: uint32_t headerBytes; src/Linux/hsflowtokens.h:HSPTOKEN_DATA( HSPTOKEN_HEADERBYTES, "headerBytes", HSPTOKENTYPE_ATTRIB ) src/Linux/hsflowconfig.c: st->headerBytes = SFL_DEFAULT_HEADER_SIZE; src/Linux/hsflowconfig.c: case HSPTOKEN_HEADERBYTES: src/Linux/hsflowconfig.c: if((tok = expectInteger32(sp, tok, &sp->sFlow->sFlowSettings_file->headerBytes, 0, HSP_MAX_HEADER_BYTES)) == NULL) return NO; With some further digging around, sflow_sampler.c -> sfl_sampler_set_sFlowFsMaximumHeaderSize(sampler, SFL_DEFAULT_HEADER_SIZE); readPackets.c -> uint32_t maxHdrLen = sampler->sFlowFsMaximumHeaderSize; It seems to read the 'headerBytes' data but not do anything with it. I also found that in Windows platform code, the parameter is (maybe?) used. samplingRate:src/Windows/hsflowd/hypervSwitch.c: sFlowConfig.sampleHeaderLength = sp->sFlow->sFlowSettings->headerBytes; Thanks, Vasanth. From: Neil McKee [mailto:nei...@in...] Sent: Tuesday, March 08, 2016 5:08 PM To: Viswanathan, Vasanth Cc: hos...@li... Subject: Re: [host-sflow-discuss] headerBytes configuration in host-sflow I checked in changes to honor both of these settings -- within the compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. I haven't tested all variations, so please let me know if you find any problems. Neil ------ Neil McKee InMon Corp. http://www.inmon.com On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth <vas...@hp...> wrote: Hi everyone, I posted a question on the sFlow google group and was redirected to this mailing list to get a better answer. I would appreciate any kind of help/clarification. 1. headerBytes configuration I am trying to configure host-sflow to use 'headerBytes' parameter. I am writing the configuration to /etc/hsflowd.conf but the packets I receive at the collector (sflowtool) do not use the configured headerBytes. For ex: headerBytes = 256, This configuration does not seem to take effect and I only see the default header bytes of 128 from sampled packets. Doing a grep on the source code shows that the 'headerBytes' parameter is read from the config file but does not seem to be used (Please correct me if I missed it) Is the functionality already supported? If not, any timelines on when we can expect it? 2. datagramBytes configuration I also plan to use 'datagramBytes' parameter to control the max datagram size. Is this already supported? Thanks for your time! Vasanth. ------------------------------------------------------------------------------ Transform Data into Opportunity. Accelerate data analysis in your applications with Intel Data Analytics Acceleration Library. Click to learn more. http://makebettercode.com/inteldaal-eval _______________________________________________ host-sflow-discuss mailing list hos...@li... https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss |
|
From: Neil M. <nei...@in...> - 2016-03-09 01:32:14
|
I checked in changes to honor both of these settings -- within the compile-time limits set in src/sflow/sflow.h and src/<platform>/hsflowd.h. I haven't tested all variations, so please let me know if you find any problems. Neil ------ Neil McKee InMon Corp. http://www.inmon.com On Tue, Mar 8, 2016 at 10:25 AM, Viswanathan, Vasanth < vas...@hp...> wrote: > Hi everyone, > > I posted a question on the sFlow google group and was redirected to this > mailing list to get a better answer. > I would appreciate any kind of help/clarification. > > 1. headerBytes configuration > I am trying to configure host-sflow to use 'headerBytes' parameter. > I am writing the configuration to /etc/hsflowd.conf but the packets I > receive at the collector (sflowtool) do not use the configured headerBytes. > For ex: headerBytes = 256, This configuration does not seem to take effect > and I only see the default header bytes of 128 from sampled packets. > Doing a grep on the source code shows that the 'headerBytes' parameter is > read from the config file but does not seem to be used (Please correct me > if I missed it) > > Is the functionality already supported? > If not, any timelines on when we can expect it? > > 2. datagramBytes configuration > I also plan to use 'datagramBytes' parameter to control the max datagram > size. > Is this already supported? > > Thanks for your time! > Vasanth. > > > ------------------------------------------------------------------------------ > Transform Data into Opportunity. > Accelerate data analysis in your applications with > Intel Data Analytics Acceleration Library. > Click to learn more. > http://makebettercode.com/inteldaal-eval > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > |
|
From: Viswanathan, V. <vas...@hp...> - 2016-03-08 18:25:33
|
Hi everyone, I posted a question on the sFlow google group and was redirected to this mailing list to get a better answer. I would appreciate any kind of help/clarification. 1. headerBytes configuration I am trying to configure host-sflow to use 'headerBytes' parameter. I am writing the configuration to /etc/hsflowd.conf but the packets I receive at the collector (sflowtool) do not use the configured headerBytes. For ex: headerBytes = 256, This configuration does not seem to take effect and I only see the default header bytes of 128 from sampled packets. Doing a grep on the source code shows that the 'headerBytes' parameter is read from the config file but does not seem to be used (Please correct me if I missed it) Is the functionality already supported? If not, any timelines on when we can expect it? 2. datagramBytes configuration I also plan to use 'datagramBytes' parameter to control the max datagram size. Is this already supported? Thanks for your time! Vasanth. |
|
From: Kevin T. <kt...@we...> - 2016-01-26 17:07:45
|
See Below: Thank you, Kevin On 1/26/16, 5:08 AM, "Sonia Panchen" <son...@in...> wrote: > >Hello Kevin, > >Please can you provide some additional information: >1. Can you clarify "This gets logged in the Event Log overtime I start the service” - do you mean that it fails immediately every time you start the service or do you mean that after some time it fails? >When I start the service it errors out within 2 - 3 mins each time. >2. Which operating system are you running and are you using the 32 or 64 bit version of hsflowd? >This machine is 2012 R2 >3. Are you running HyperV and if so can you describe the configuration of the vms? If not can you describe the configuration of the system? Yes I am running HyperV. >Some of these devices have Legacy Network adapters some are the standard NIC, all of them are using VLAN tagging. >4. Please run hsflowd.exe from the command line with -vv option (see http://www.sflow.net/host-sflow-win-config.php section on logging) and send the output. >Same result as when I start with the service. I have attached the log1.txt file which is the result of the verbose output. I will work on the debug stuff for you this afternoon. > >To debug this further it would be useful if you can generate a process dump of the crash. To do this you will need to create a registry key and value (from >https://www.novell.com/support/kb/doc.php?id=7013369 <https://www.novell.com/support/kb/doc.php?id=7013369>;): > >[HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\Windows Error Reporting\LocalDumps] >"DumpType"=dword:00000002 > >Note the "LocalDumps" subkey typically does not exist by default, and must be created in addition to the actual "DumpType" DWORD value within that subkey. > >From Powershell as Administrator, then run hsflowd.exe. This should generate a .dmp in "%LOCALAPPDATA%\CrashDumps” directory relative to your user profile. Please then send me the .dmp (I will contact you directly to arrange this). > >Best regards, >Sonia Panchen > >> On 25 Jan 2016, at 23:59, Kevin Tucker <kt...@we...> wrote: >> >> >> This gets logged in the Event Log overtime I start the service. >> >> Log Name: Application >> Source: Application Error >> Date: 1/25/2016 4:40:05 PM >> Event ID: 1000 >> Task Category: (100) >> Level: Error >> Keywords: Classic >> User: N/A >> Computer: >> Description: >> Faulting application name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a >> Faulting module name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a >> Exception code: 0xc0000409 >> Fault offset: 0x0000000000019abf >> Faulting process id: 0x278c >> Faulting application start time: 0x01d157c13d16aa01 >> Faulting application path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe >> Faulting module path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe >> Report Id: 93978387-c3b4-11e5-9437-842b2b0227c4 >> Faulting package full name: >> Faulting package-relative application ID: >> Event Xml: >> <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event"> >> <System> >> <Provider Name="Application Error" /> >> <EventID Qualifiers="0">1000</EventID> >> <Level>2</Level> >> <Task>100</Task> >> <Keywords>0x80000000000000</Keywords> >> <TimeCreated SystemTime="2016-01-25T22:40:05.000000000Z" /> >> <EventRecordID>250682</EventRecordID> >> <Channel>Application</Channel> >> <Computer></Computer> >> <Security /> >> </System> >> <EventData> >> <Data>hsflowd.exe</Data> >> <Data>1.23.5.0</Data> >> <Data>55683c4a</Data> >> <Data>hsflowd.exe</Data> >> <Data>1.23.5.0</Data> >> <Data>55683c4a</Data> >> <Data>c0000409</Data> >> <Data>0000000000019abf</Data> >> <Data>278c</Data> >> <Data>01d157c13d16aa01</Data> >> <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> >> <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> >> <Data>93978387-c3b4-11e5-9437-842b2b0227c4</Data> >> <Data> >> </Data> >> <Data> >> </Data> >> </EventData> >> </Event> >> >> >> Thank you, >> >> Kevin Tucker >> >> ________________________________ >> >> Weatherford, Texas is a service oriented, yet still "small town" community that upgrades its infrastructure, celebrates and shares its rich history, and fosters quality economic growth-in a fiscally-responsible, even revenue-generating fashion. >> >> This message (and any associated files) is intended only for the use of the individual or entity to which it is addressed and may contain information that is confidential, or subject to copyright. If you are not the intended recipient you are hereby notified that any dissemination, copying or distribution of this message, or files associated with this message, is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from the City of Weatherford may be monitored. >> >> Internet communications cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. Therefore, we do not accept responsibility for any errors or omissions that are present in this message, or any attachment, that have arisen as a result of e-mail transmission. If verification is required, please request a hard-copy version. Any views or opinions presented are solely those of the author and do not necessarily represent those of the City of Weatherford. >> >> ------------------------------------------------------------------------------ >> Site24x7 APM Insight: Get Deep Visibility into Application Performance >> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month >> Monitor end-to-end web transactions and take corrective actions now >> Troubleshoot faster and improve end-user experience. Signup Now! >> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 >> _______________________________________________ >> host-sflow-discuss mailing list >> hos...@li... >> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss > |
|
From: Sonia P. <son...@in...> - 2016-01-26 11:33:59
|
Hello Kevin, Please can you provide some additional information: 1. Can you clarify "This gets logged in the Event Log overtime I start the service” - do you mean that it fails immediately every time you start the service or do you mean that after some time it fails? 2. Which operating system are you running and are you using the 32 or 64 bit version of hsflowd? 3. Are you running HyperV and if so can you describe the configuration of the vms? If not can you describe the configuration of the system? 4. Please run hsflowd.exe from the command line with -vv option (see http://www.sflow.net/host-sflow-win-config.php section on logging) and send the output. To debug this further it would be useful if you can generate a process dump of the crash. To do this you will need to create a registry key and value (from https://www.novell.com/support/kb/doc.php?id=7013369 <https://www.novell.com/support/kb/doc.php?id=7013369>;): [HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\Windows Error Reporting\LocalDumps] "DumpType"=dword:00000002 Note the "LocalDumps" subkey typically does not exist by default, and must be created in addition to the actual "DumpType" DWORD value within that subkey. From Powershell as Administrator, then run hsflowd.exe. This should generate a .dmp in "%LOCALAPPDATA%\CrashDumps” directory relative to your user profile. Please then send me the .dmp (I will contact you directly to arrange this). Best regards, Sonia Panchen > On 25 Jan 2016, at 23:59, Kevin Tucker <kt...@we...> wrote: > > > This gets logged in the Event Log overtime I start the service. > > Log Name: Application > Source: Application Error > Date: 1/25/2016 4:40:05 PM > Event ID: 1000 > Task Category: (100) > Level: Error > Keywords: Classic > User: N/A > Computer: > Description: > Faulting application name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a > Faulting module name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a > Exception code: 0xc0000409 > Fault offset: 0x0000000000019abf > Faulting process id: 0x278c > Faulting application start time: 0x01d157c13d16aa01 > Faulting application path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe > Faulting module path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe > Report Id: 93978387-c3b4-11e5-9437-842b2b0227c4 > Faulting package full name: > Faulting package-relative application ID: > Event Xml: > <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event"> > <System> > <Provider Name="Application Error" /> > <EventID Qualifiers="0">1000</EventID> > <Level>2</Level> > <Task>100</Task> > <Keywords>0x80000000000000</Keywords> > <TimeCreated SystemTime="2016-01-25T22:40:05.000000000Z" /> > <EventRecordID>250682</EventRecordID> > <Channel>Application</Channel> > <Computer></Computer> > <Security /> > </System> > <EventData> > <Data>hsflowd.exe</Data> > <Data>1.23.5.0</Data> > <Data>55683c4a</Data> > <Data>hsflowd.exe</Data> > <Data>1.23.5.0</Data> > <Data>55683c4a</Data> > <Data>c0000409</Data> > <Data>0000000000019abf</Data> > <Data>278c</Data> > <Data>01d157c13d16aa01</Data> > <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> > <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> > <Data>93978387-c3b4-11e5-9437-842b2b0227c4</Data> > <Data> > </Data> > <Data> > </Data> > </EventData> > </Event> > > > Thank you, > > Kevin Tucker > > ________________________________ > > Weatherford, Texas is a service oriented, yet still "small town" community that upgrades its infrastructure, celebrates and shares its rich history, and fosters quality economic growth-in a fiscally-responsible, even revenue-generating fashion. > > This message (and any associated files) is intended only for the use of the individual or entity to which it is addressed and may contain information that is confidential, or subject to copyright. If you are not the intended recipient you are hereby notified that any dissemination, copying or distribution of this message, or files associated with this message, is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from the City of Weatherford may be monitored. > > Internet communications cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. Therefore, we do not accept responsibility for any errors or omissions that are present in this message, or any attachment, that have arisen as a result of e-mail transmission. If verification is required, please request a hard-copy version. Any views or opinions presented are solely those of the author and do not necessarily represent those of the City of Weatherford. > > ------------------------------------------------------------------------------ > Site24x7 APM Insight: Get Deep Visibility into Application Performance > APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month > Monitor end-to-end web transactions and take corrective actions now > Troubleshoot faster and improve end-user experience. Signup Now! > http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss |
|
From: Kevin T. <kt...@we...> - 2016-01-25 23:59:41
|
This gets logged in the Event Log overtime I start the service. Log Name: Application Source: Application Error Date: 1/25/2016 4:40:05 PM Event ID: 1000 Task Category: (100) Level: Error Keywords: Classic User: N/A Computer: Description: Faulting application name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a Faulting module name: hsflowd.exe, version: 1.23.5.0, time stamp: 0x55683c4a Exception code: 0xc0000409 Fault offset: 0x0000000000019abf Faulting process id: 0x278c Faulting application start time: 0x01d157c13d16aa01 Faulting application path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe Faulting module path: C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe Report Id: 93978387-c3b4-11e5-9437-842b2b0227c4 Faulting package full name: Faulting package-relative application ID: Event Xml: <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event"> <System> <Provider Name="Application Error" /> <EventID Qualifiers="0">1000</EventID> <Level>2</Level> <Task>100</Task> <Keywords>0x80000000000000</Keywords> <TimeCreated SystemTime="2016-01-25T22:40:05.000000000Z" /> <EventRecordID>250682</EventRecordID> <Channel>Application</Channel> <Computer></Computer> <Security /> </System> <EventData> <Data>hsflowd.exe</Data> <Data>1.23.5.0</Data> <Data>55683c4a</Data> <Data>hsflowd.exe</Data> <Data>1.23.5.0</Data> <Data>55683c4a</Data> <Data>c0000409</Data> <Data>0000000000019abf</Data> <Data>278c</Data> <Data>01d157c13d16aa01</Data> <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> <Data>C:\Program Files\Host sFlow Project\Host sFlow Agent\hsflowd.exe</Data> <Data>93978387-c3b4-11e5-9437-842b2b0227c4</Data> <Data> </Data> <Data> </Data> </EventData> </Event> Thank you, Kevin Tucker ________________________________ Weatherford, Texas is a service oriented, yet still "small town" community that upgrades its infrastructure, celebrates and shares its rich history, and fosters quality economic growth-in a fiscally-responsible, even revenue-generating fashion. This message (and any associated files) is intended only for the use of the individual or entity to which it is addressed and may contain information that is confidential, or subject to copyright. If you are not the intended recipient you are hereby notified that any dissemination, copying or distribution of this message, or files associated with this message, is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from the City of Weatherford may be monitored. Internet communications cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. Therefore, we do not accept responsibility for any errors or omissions that are present in this message, or any attachment, that have arisen as a result of e-mail transmission. If verification is required, please request a hard-copy version. Any views or opinions presented are solely those of the author and do not necessarily represent those of the City of Weatherford. |
|
From: Sonia P. <son...@in...> - 2015-11-03 15:44:20
|
Brahim, I had in mind that you could install sflowtool on each Win 2003 Server. As you point out this avoids SPOF. The blog post is useful for guidance on how to run sflowtool with forwarding - the instructions are the same for both a central forwarder or a forwarder per server. Sonia > On 3 Nov 2015, at 11:50, Brahim EL Hachimi <bel...@ho...> wrote: > > Hi Sonia, > > Thank you for your quick answer. This is a good catch and i've been looking at the sFlowtool for a while. > But the solution i've been trying to set up is the one presented on the top left on the Forwarding using sflowtool <http://blog.sflow.com/2012/01/forwarding-using-sflowtool.html> . Indeed my need is to have each host sFlow agent installed on each win 2003 server sending to two Ganglia gmond master servers (eg could be considered as sFlow analyzers though). I do want to avoid adding a forwarder in between because it will add a SPOF (if it fails). > My understanding then, is that i should install the sFlowtool on each of my win2K3 servers, and use it to forward localhost host sFlow datagrams to my two gmond masters ? > > Is this the idea you use to have ? > > Thank you. > > > > Subject: Re: [host-sflow-discuss] host sFlow configuration issue on Windows 2003 > From: son...@in... <mailto:son...@in...> > Date: Tue, 3 Nov 2015 09:41:16 +0000 > CC: hos...@li... <mailto:hos...@li...> > To: bel...@ho... <mailto:bel...@ho...> > > The Windows XP version of hsflowd supports only one collector. > > Another alternative, to work around this, is to use sflowtool (http://www.inmon.com/technology/sflowTools.php <http://www.inmon.com/technology/sflowTools.php>) to forward to multiple collectors (see Forwarding using sflowtool <http://blog.sflow.com/2012/01/forwarding-using-sflowtool.html> for an example). You would probably need to set this up as a service. > > Best regards, > Sonia Panchen > > > On 2 Nov 2015, at 20:44, Brahim EL Hachimi <bel...@ho... <mailto:bel...@ho...>> wrote: > > Hi gents, > > Sorry to bother, but i wasn't able to grab the answer to my question anywhere. I'm fighting since few days to make my hsflowd-winxp-<version>-x86.msi <https://github.com/sflow/host-sflow/releases/download/REL-1_23_3/hsflowd-winxp-1.23.3-x86.msi> installed on win2003 servers, sending to two collectors. (because i've switched to a redundant Ganglia infrastructures with 2 gmond nodes) > > Although i've seen (and tested) this is possible on x64 and also on x86 for Vista & upward systems, the registry "tweak" described here <http://www.sflow.net/host-sflow-win-config.php> isn't operational for the above agent version. > > May i kindly ask you, if you've a solution ? I've found a workaround which consists of installing two instances of sFlowd (sFlowd & sFlowd2 with associated services) but i doesn't look to be a professional and clean solution. > > Thank you so much for your help. > ------------------------------------------------------------------------------ > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... <mailto:hos...@li...> > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss <https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss> |
|
From: Brahim EL H. <bel...@ho...> - 2015-11-03 11:51:07
|
Hi Sonia, Thank you for your quick answer. This is a good catch and i've been looking at the sFlowtool for a while. But the solution i've been trying to set up is the one presented on the top left on the Forwarding using sflowtool . Indeed my need is to have each host sFlow agent installed on each win 2003 server sending to two Ganglia gmond master servers (eg could be considered as sFlow analyzers though). I do want to avoid adding a forwarder in between because it will add a SPOF (if it fails). My understanding then, is that i should install the sFlowtool on each of my win2K3 servers, and use it to forward localhost host sFlow datagrams to my two gmond masters ? Is this the idea you use to have ? Thank you. Subject: Re: [host-sflow-discuss] host sFlow configuration issue on Windows 2003 From: son...@in... Date: Tue, 3 Nov 2015 09:41:16 +0000 CC: hos...@li... To: bel...@ho... The Windows XP version of hsflowd supports only one collector. Another alternative, to work around this, is to use sflowtool (http://www.inmon.com/technology/sflowTools.php) to forward to multiple collectors (see Forwarding using sflowtool for an example). You would probably need to set this up as a service. Best regards,Sonia Panchen On 2 Nov 2015, at 20:44, Brahim EL Hachimi <bel...@ho...> wrote:Hi gents, Sorry to bother, but i wasn't able to grab the answer to my question anywhere. I'm fighting since few days to make my hsflowd-winxp-<version>-x86.msi installed on win2003 servers, sending to two collectors. (because i've switched to a redundant Ganglia infrastructures with 2 gmond nodes) Although i've seen (and tested) this is possible on x64 and also on x86 for Vista & upward systems, the registry "tweak" described here isn't operational for the above agent version. May i kindly ask you, if you've a solution ? I've found a workaround which consists of installing two instances of sFlowd (sFlowd & sFlowd2 with associated services) but i doesn't look to be a professional and clean solution. Thank you so much for your help. ------------------------------------------------------------------------------_______________________________________________host-sflow-discuss mailing lis...@li...://lists.sourceforge.net/lists/listinfo/host-sflow-discuss |
|
From: Sonia P. <son...@in...> - 2015-11-03 10:11:47
|
The Windows XP version of hsflowd supports only one collector. Another alternative, to work around this, is to use sflowtool (http://www.inmon.com/technology/sflowTools.php <http://www.inmon.com/technology/sflowTools.php>) to forward to multiple collectors (see Forwarding using sflowtool <http://blog.sflow.com/2012/01/forwarding-using-sflowtool.html> for an example). You would probably need to set this up as a service. Best regards, Sonia Panchen > On 2 Nov 2015, at 20:44, Brahim EL Hachimi <bel...@ho...> wrote: > > Hi gents, > > Sorry to bother, but i wasn't able to grab the answer to my question anywhere. I'm fighting since few days to make my hsflowd-winxp-<version>-x86.msi <https://github.com/sflow/host-sflow/releases/download/REL-1_23_3/hsflowd-winxp-1.23.3-x86.msi> installed on win2003 servers, sending to two collectors. (because i've switched to a redundant Ganglia infrastructures with 2 gmond nodes) > > Although i've seen (and tested) this is possible on x64 and also on x86 for Vista & upward systems, the registry "tweak" described here <http://www.sflow.net/host-sflow-win-config.php> isn't operational for the above agent version. > > May i kindly ask you, if you've a solution ? I've found a workaround which consists of installing two instances of sFlowd (sFlowd & sFlowd2 with associated services) but i doesn't look to be a professional and clean solution. > > Thank you so much for your help. > ------------------------------------------------------------------------------ > _______________________________________________ > host-sflow-discuss mailing list > hos...@li... <mailto:hos...@li...> > https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss <https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss> |
|
From: Brahim EL H. <bel...@ho...> - 2015-11-02 20:45:06
|
Hi gents, Sorry to bother, but i wasn't able to grab the answer to my question anywhere. I'm fighting since few days to make my hsflowd-winxp-<version>-x86.msi installed on win2003 servers, sending to two collectors. (because i've switched to a redundant Ganglia infrastructures with 2 gmond nodes) Although i've seen (and tested) this is possible on x64 and also on x86 for Vista & upward systems, the registry "tweak" described here isn't operational for the above agent version. May i kindly ask you, if you've a solution ? I've found a workaround which consists of installing two instances of sFlowd (sFlowd & sFlowd2 with associated services) but i doesn't look to be a professional and clean solution. Thank you so much for your help. |
|
From: Neil M. <nei...@in...> - 2015-09-17 18:16:02
|
Hello all, Please note that the new home page for host-sflow is http://sflow.net, and the source code is now on github at https://github.com/sflow/host-sflow. Neil ------ Neil McKee InMon Corp. http://www.inmon.com |
|
From: Neil M. <nei...@in...> - 2015-07-27 15:51:45
|
There is now a 1.28.1 release with a XenServer 6.5 binary here: https://github.com/sflow/host-sflow/releases I believe more work is needed to fill in all the VM stats, but please test it and let me know how it goes. This switch from sourceforge to github may just be a temporary measure, but if sourceforce continues to be unstable then we may make it permanent. Regards, Neil ------ Neil McKee InMon Corp. http://www.inmon.com On Thu, Jul 9, 2015 at 3:04 PM, Neil McKee <nei...@in...> wrote: > I downloaded the DDK .iso from > http://xenserver.org/open-source-virtualization-download.html and > installed it on a xenserver 6.5 (SP1) server according to the CLI > instructions at > https://support.citrix.com/servlet/KbServlet/download/20640-102-489043/ddk-1.0-5.5.0-en_gb.pdf > . > > hsflowd compiled OK (with a small tweak to xenserver-ddk/Makefile to set > ARCH to x86_64). It reports the hypervisor stats and configures sFlow in > openvswitch OK but it doesn't seem to be picking up the per-VM cpu/mem/io > stats, so that API must have changed. I'll reply here again when it's > fixed and the xenserver 6.5 binary is posted to the downloads list on > http://host-sflow.sf.net. > > Neil > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Thu, Jul 9, 2015 at 2:35 PM, Elias Abacioglu <eli...@gm...> > wrote: > >> Where can I get this DDK VM? >> >> 2015-07-09 21:07 GMT+02:00 Neil McKee <nei...@in...>: >> >>> I'll take a look. >>> >>> (but if you have the DDK VM installed on your 6.5 cluster somewhere then >>> you could try running "make xenserver" on the hsflowd sources. That's all >>> we do for the other releases.) >>> >>> Neil >>> >>> >>> ------ >>> Neil McKee >>> InMon Corp. >>> http://www.inmon.com >>> >>> On Thu, Jul 9, 2015 at 11:47 AM, Elias Abacioglu <eli...@gm...> >>> wrote: >>> >>>> Hi, >>>> >>>> Is there any plan to package Host sflow to XenServer 6.5? >>>> It's been out for a while now. >>>> Basically the difference from 6.2 is that it's x86_64 architecture and >>>> it's based on CentOS 5.10 I think. >>>> >>>> /Elias >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> Don't Limit Your Business. Reach for the Cloud. >>>> GigeNET's Cloud Solutions provide you with the tools and support that >>>> you need to offload your IT needs and focus on growing your business. >>>> Configured For All Businesses. Start Your Cloud Today. >>>> https://www.gigenetcloud.com/ >>>> _______________________________________________ >>>> host-sflow-discuss mailing list >>>> hos...@li... >>>> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss >>>> >>>> >>> >> > |
|
From: Jean-Baptiste D. <jb...@pa...> - 2015-07-10 10:32:29
|
On 07/08/2015 07:30 PM, Neil McKee wrote: > Yes, for now that feature is only in the Linux port. > > It's an implementation of this part of the sFlow standard: > > http://sflow.org/sflow_application.txt > > which defines an abstract model for measuring client-server transactions. For > some protocols there are more specific structures (eg. HTML, memcached) but this > generic extension can be used for almost anything. Like other sFlow > measurements it combines a core set of counters with 1-:N sampled transactions. > So the core metrics are counted exactly, but you can derive almost any other > metric on a sampled basis. OK, thank you for the answer. Jean-Baptiste |
|
From: Neil M. <nei...@in...> - 2015-07-09 22:04:41
|
I downloaded the DDK .iso from http://xenserver.org/open-source-virtualization-download.html and installed it on a xenserver 6.5 (SP1) server according to the CLI instructions at https://support.citrix.com/servlet/KbServlet/download/20640-102-489043/ddk-1.0-5.5.0-en_gb.pdf . hsflowd compiled OK (with a small tweak to xenserver-ddk/Makefile to set ARCH to x86_64). It reports the hypervisor stats and configures sFlow in openvswitch OK but it doesn't seem to be picking up the per-VM cpu/mem/io stats, so that API must have changed. I'll reply here again when it's fixed and the xenserver 6.5 binary is posted to the downloads list on http://host-sflow.sf.net. Neil ------ Neil McKee InMon Corp. http://www.inmon.com On Thu, Jul 9, 2015 at 2:35 PM, Elias Abacioglu <eli...@gm...> wrote: > Where can I get this DDK VM? > > 2015-07-09 21:07 GMT+02:00 Neil McKee <nei...@in...>: > >> I'll take a look. >> >> (but if you have the DDK VM installed on your 6.5 cluster somewhere then >> you could try running "make xenserver" on the hsflowd sources. That's all >> we do for the other releases.) >> >> Neil >> >> >> ------ >> Neil McKee >> InMon Corp. >> http://www.inmon.com >> >> On Thu, Jul 9, 2015 at 11:47 AM, Elias Abacioglu <eli...@gm...> >> wrote: >> >>> Hi, >>> >>> Is there any plan to package Host sflow to XenServer 6.5? >>> It's been out for a while now. >>> Basically the difference from 6.2 is that it's x86_64 architecture and >>> it's based on CentOS 5.10 I think. >>> >>> /Elias >>> >>> >>> ------------------------------------------------------------------------------ >>> Don't Limit Your Business. Reach for the Cloud. >>> GigeNET's Cloud Solutions provide you with the tools and support that >>> you need to offload your IT needs and focus on growing your business. >>> Configured For All Businesses. Start Your Cloud Today. >>> https://www.gigenetcloud.com/ >>> _______________________________________________ >>> host-sflow-discuss mailing list >>> hos...@li... >>> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss >>> >>> >> > |
|
From: Elias A. <eli...@gm...> - 2015-07-09 21:35:13
|
Where can I get this DDK VM? 2015-07-09 21:07 GMT+02:00 Neil McKee <nei...@in...>: > I'll take a look. > > (but if you have the DDK VM installed on your 6.5 cluster somewhere then > you could try running "make xenserver" on the hsflowd sources. That's all > we do for the other releases.) > > Neil > > > ------ > Neil McKee > InMon Corp. > http://www.inmon.com > > On Thu, Jul 9, 2015 at 11:47 AM, Elias Abacioglu <eli...@gm...> > wrote: > >> Hi, >> >> Is there any plan to package Host sflow to XenServer 6.5? >> It's been out for a while now. >> Basically the difference from 6.2 is that it's x86_64 architecture and >> it's based on CentOS 5.10 I think. >> >> /Elias >> >> >> ------------------------------------------------------------------------------ >> Don't Limit Your Business. Reach for the Cloud. >> GigeNET's Cloud Solutions provide you with the tools and support that >> you need to offload your IT needs and focus on growing your business. >> Configured For All Businesses. Start Your Cloud Today. >> https://www.gigenetcloud.com/ >> _______________________________________________ >> host-sflow-discuss mailing list >> hos...@li... >> https://lists.sourceforge.net/lists/listinfo/host-sflow-discuss >> >> > |
2 messages has been excluded from this view by a project administrator.
