sflow Source stops working

[DennisPR]

We have installed VMWare tools on the VM and upgraded it from 1 to 4GB but the issue remains

[tgriep]

A new version of Network Analyzer was released last week. Can you upgrade to that and see it the issue still happens?
http://www.nagios.com/news/company-news ... 0-released

[DennisPR]

We are using version 2R1.0 and still have the issue

[tgriep]

Could you post how the Hp Flex virtual network (sFlow) device is setup?
Can give us a make and model number of that device too?

[Box293]

I would like to see how much free disk space there is:

Type the following at the command prompt:

Code: Select all

df -h
dh -i

Please post that output here.

If disk usage is your problem, run these commands to product a breakdown of the directories at / (NOTE: This may take some time to finish).

Code: Select all

cd /
du -sh *

If you want to see more detail about a directory, cd into it and run the du -sh * command again.

[DennisPR]

Code: Select all

[root@AP-DCO163-NA ~]# df -h
Filesystem      Size  Used Avail Use% Mounted on
rootfs          9.4G  3.4G  6.0G  36% /
devtmpfs        2.0G  140K  2.0G   1% /dev
tmpfs           2.0G     0  2.0G   0% /dev/shm
/dev/sda1       9.4G  3.4G  6.0G  36% /
[root@AP-DCO163-NA ~]# dh -i
-bash: dh: command not found
[root@AP-DCO163-NA ~]# df -i
Filesystem     Inodes IUsed  IFree IUse% Mounted on
rootfs         625856 47421 578435    8% /
devtmpfs       504856   565 504291    1% /dev
tmpfs          507026     1 507025    1% /dev/shm
/dev/sda1      625856 47421 578435    8% /
[root@AP-DCO163-NA ~]# cd /
[root@AP-DCO163-NA /]# du -sh *
5.8M    bin
43M     boot
140K    dev
29M     etc
20K     home
251M    lib
18M     lib64
16K     lost+found
4.0K    media
4.0K    mnt
4.0K    opt
du: cannot access `proc/14822/task/14822/fd/4': No such file or directory
du: cannot access `proc/14822/task/14822/fdinfo/4': No such file or directory
du: cannot access `proc/14822/fd/4': No such file or directory
du: cannot access `proc/14822/fdinfo/4': No such file or directory
0       proc
52K     root
9.1M    sbin
4.0K    selinux
4.0K    srv
256M    swapfile1
0       sys
344M    tmp
1.6G    usr
729M    var

[jdalrymple]

Could you post how the Hp Flex virtual network (sFlow) device is setup?
Can give us a make and model number of that device too?

In addition to this, I wonder what version of sflow the HP switch is spewing out. Maybe that information is in what tgriep is requesting. Unfortunately the documentation that I'm seeing for the HP VC switches doesn't offer a whole lot of insight

*it won't help* but maybe a step to take would be to dump the data to a pcap file then feed it into sfcapd in a more "labby" type environment then maybe more data can be extracted from sfcapd during the crash. Unfortunately you seem to have run across a pretty ungoogleable problem. As a matter of fact, you're #2 when googling "sfcapd crashes"

[nagios-retail]

In attachement you can find info about the Flex card & config :

best regards
Koen Beuselinck

[tgriep]

Thanks for the Info. I went through the manual and there isn't much to edit in there.
You may want to change the Max Header and Datagram size to see if it helps.
Couple of things to try.
When the flow stops in NA, can you run a tcpdump on the NA system to see if it is still receiving data from the HP Switch.
Also, go to the Sflow Status screen on the HP and see if there are any errors there.
Let us know what you find.

[nagios-retail]

Do you have suggestions to adjust these settings (Max header & Diagram size) ? We used the default settings.
As mentioned in an earlier post before we keep on receiving flows from HP.