<div dir="auto">Maybe a good reminder that you can also bring a tcpdump file (-w) into Wireshark for a GUI based analysis. <div dir="auto"><br></div><div dir="auto">Also I've seen IBM support do a lot with a packet capture on a ticket, as well as a snap.</div><div dir="auto"><br></div><div dir="auto">Alec</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Aug 29, 2023, 2:17 AM Helge Hauglin <<a href="mailto:helge.hauglin@usit.uio.no">helge.hauglin@usit.uio.no</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
To identify which address sends most packages to and from a protocol<br>
node, I use a variation of this:<br>
<br>
| tcpdump -c 20000 -i <interface> 2>/dev/null | grep IP | cut -d' ' -f3 | sort | uniq -c | sort -nr | head -10<br>
<br>
(Collect 20.000 packages, pick out sender address and port, sort and<br>
count those, make a top 10 list.)<br>
<br>
You could limit to only NFS traffic by adding "port nfs" at the end of<br>
the "tcpdump" command, but then you would not see e.g SMB clients with a<br>
lot of traffic, if there are any of those.<br>
<br>
> Hallo,<br>
><br>
> since some time we do have seemingly random issues with a particular<br>
> customer accessing data over Ganesha / CES (5.1.8). What happens is<br>
> that the CES server owning their IP gets a very high cpu load, and<br>
> every operation on the NFS clients become sluggish. It does seem not<br>
> related to throughput, and looking at the metrics [*] I do not see a<br>
> correlation with e.g. increased NFS ops. I see no events in GPFS, and<br>
> nothing suspicious in the ganesha and gpfs log files.<br>
><br>
> What would be a good procedure to identify the misbehaving client (I<br>
> suspect NFS, as it seems there is only 1 idle SMB client)? I have put<br>
> now LOGLEVEL=INFO in ganesha to see if I catch anything interesting,<br>
> but I would be curious on how this kind of apparently random issues<br>
> could be better debugged and restricted to a client<br>
><br>
> Thanks a lot!<br>
><br>
> regards<br>
><br>
> leo<br>
><br>
> [*]<br>
><br>
> for i in read write; do for j in ops queue lat req err; do mmperfmon<br>
> query "ces-server|NFSIO|/export/path|NFSv41|nfs_${i}_$j"<br>
> 2023-08-25-14:40:00 2023-08-25-15:05:00 -b60; done; done<br>
<br>
-- <br>
Regards,<br>
<br>
Helge Hauglin<br>
<br>
----------------------------------------------------------------<br>
Mr. Helge Hauglin, Senior Engineer<br>
System administrator<br>
Center for Information Technology, University of Oslo, Norway<br>
<br>
_______________________________________________<br>
gpfsug-discuss mailing list<br>
gpfsug-discuss at <a href="http://gpfsug.org" rel="noreferrer noreferrer" target="_blank">gpfsug.org</a><br>
<a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org" rel="noreferrer noreferrer" target="_blank">http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org</a><br>
</blockquote></div>