[Check_mk (english)] check_mk enable debug logging
Aravind.Jangam at proquest.com
Thu Feb 28 23:49:34 CET 2019
Thanks for the clarification and details.
We had a query. We are using check_mk to monitor certain AWS cloud components like EFS which has been mounted on each of the EC2 instances where check_mk agent is running. We are getting this stale handle critical alert while monitoring the EFS mounts. Since we do not have any performance issues shown via cloud watch stats or complaints by any application owners that they are not able to access the data residing on these mounts, we were trying to explore the root cause of these alerts.
Are we missing any configuration for check_mk which would cause this to happen since we are trying to monitor components in AWS instead of on-prem counterparts? Could you please guide us?
From: Paul Dott <pauldott at gmail.com>
Sent: Thursday, February 28, 2019 11:20 AM
To: Aravind Jangam <Aravind.Jangam at proquest.com>
Cc: checkmk-en at lists.mathias-kettner.de; Ruchi Saxena <Ruchi.Saxena at proquest.com>
Subject: Re: [Check_mk (english)] check_mk enable debug logging
I don't think increasing the logging will help here. This stale handle is really a result of the 'waitmax' program kicking in.
Section from the agent;
sed -n '/ nfs4\? /s/[^ ]* \([^ ]*\) .*/\1/p' < /proc/mounts |
sed 's/\\040/ /g' |
while read MP
if [ "$STAT_VERSION" != "$STAT_BROKE" ]; then
waitmax -s 9 5 stat -f -c "$MP ok %b %f %a %s" "$MP" || \
echo "$MP hanging 0 0 0 0"
waitmax -s 9 5 stat -f -c "$MP ok %b %f %a %s" "$MP" && \
printf '\n'|| echo "$MP hanging 0 0 0 0"
The agent tried to query the mount points, and if not responding, then it causes this response (essentially no data). This is described a little better here - https://mathias-kettner.de/cms_check_nfsmounts.html<https://urldefense.proofpoint.com/v2/url?u=https-3A__mathias-2Dkettner.de_cms-5Fcheck-5Fnfsmounts.html&d=DwMFaQ&c=WMhnfwkfN4LR6wX29ZSgFCZf_hw4vy5MAv7iZJNaAD4&r=s5797Q2TauRaAiLXpG1-pMzG5gVYnAILS2OawDN33z4&m=dQ3OE7VatdYrrCmebPJ9619iCBIRVK4fzmM3zl5olY0&s=8SuWuV5B31kul5J8395Rf3ptrjXRGib3iI9ogW6hwuM&e=>
On Thu, Feb 28, 2019 at 8:25 AM Aravind Jangam <Aravind.Jangam at proquest.com<mailto:Aravind.Jangam at proquest.com>> wrote:
Couple of questions
1. I wanted to enable debug logging for check_mk
I added below line in /etc/check_mk/main.mk<https://urldefense.proofpoint.com/v2/url?u=http-3A__main.mk&d=DwMFaQ&c=WMhnfwkfN4LR6wX29ZSgFCZf_hw4vy5MAv7iZJNaAD4&r=s5797Q2TauRaAiLXpG1-pMzG5gVYnAILS2OawDN33z4&m=dQ3OE7VatdYrrCmebPJ9619iCBIRVK4fzmM3zl5olY0&s=2KhcQa0HXEJap8PdsOYRCX3Wb0fBnW8jRW32_n_y3Ek&e=> file and restarted check_mk (cmk -R)
debug_log = "/usr/local/nagios/var/check_mk_debug.log"
Still, nothing is being logged, can you please let me know how to achieve this
1. Some of the NFS mounts keep failing with "Stale fs handle"
I was looking at check_mk scripts for nfsmounts (/usr/share/check_mk/checks/nfsmounts & /usr/share/check_mk/checks/network_fs.include)
Please find them attached
When looking at network_fs.include, below are the conditions for Stale fs handle
How can we find out which of these conditions is causing this state (OR) find out what this casing stale fs handle error
if size_blocks <= 0 or free_blocks < 0 or blocksize > 1024*1024:
return 2, "Stale fs handle"
checkmk-en mailing list
checkmk-en at lists.mathias-kettner.de<mailto:checkmk-en at lists.mathias-kettner.de>
Manage your subscription or unsubscribe
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the checkmk-en