IOError Bad file descriptor

checkmk
error
wato

(Leon) #1

Good morning!

Today I got an error when I renamed a host. In WATO I get the following output in the Background Job Details. The renaming was successful, but I still find this questionable…

Renaming WATO configuration...
Renaming host(s) in folders...
Renaming host(s) in cluster nodes...
Renaming host(s) in parents...
Renaming host(s) in rulesets...
Renaming host(s) in BI aggregations...
Renaming host(s) in base configuration, rrd, history files, etc.
This might take some time and involves a core restart...
Process Process-225:1:1:
Traceback (most recent call last):
  File "/omd/sites/master/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    sys.stdin.close()
IOError: [Errno 9] Bad file descriptor
Renaming host(s) in notification rules...
Calling final hooks
Master Disk IO Summary

OK - Utilization: 0.1%, Read: 0.00 B/s, Write: 196.33 kB/s, Average Wait: 0.06 ms, Average Read Wait: 0.00 ms, Average Write Wait: 0.06 ms, Latency: 0.06 ms, Average Queue Length: 0.00

Master Filesystem

OK - 19.7% used (6.97 of 35.31 GB), trend: +16.93 MB / 24 hours

Slave Disk IO Summary

OK - Utilization: 0.1%, Read: 68.27 B/s, Write: 193.33 kB/s, Average Wait: 0.09 ms, Average Read Wait: 4.00 ms, Average Write Wait: 0.09 ms, Latency: 0.09 ms, Average Queue Length: 0.00

Slave Filesystem

OK - 19.8% used (6.98 of 35.31 GB), trend: +15.00 MB / 24 hours

System Information

check_mk version Master & Slave: 1.5.0p6
Master: Ubuntu 16.04.5 LTS
Slave with hosts: Ubuntu 16.04.5 LTS

I thank you in advance for any approach to correct this error.

Cheers
Leon :v:t3:


#2

did you try : cmk -R as site user?


(Leon) #3

Now i did it on my master, here is the result: :smile:

OMD[master]:~$ cmk -R
Generating configuration for core (type nagios)...OK
Validating Nagios configuration...OK
Precompiling host checks...OK
Restarting monitoring core...OK 

FYI: I have on my master one configured site (master) and configure my own hosts over distributed WATO

Master (no hosts, except localhost)
- sig: hosts of my network
- slave2: customerA site with customerA hosts
- slave3: customerB site with customerB hosts

The host i have changed was part of slave “sig”.


#4

you have to do that on the site where the host is created :slight_smile:


(Leon) #5

Okay Neeloj,

OMD[sig]:~$ cmk -R
Generating configuration for core (type nagios)...OK
Validating Nagios configuration...OK
Precompiling host checks...OK
Restarting monitoring core...OK

This is the output on the internal slave, on which the host was created.


#6

and you still have ?


(Leon) #7

Yes, I renamed another host and just got the same ouput at WATO.


#8

usually chekc_mk-Team dont recommend that to rename hosts.
if you delete it and create it again its better :slight_smile:


(Leon) #9

The thing is to install the function but then recommend not to use it… :rofl:

Well, then I’ll try to stick to your tip.
I hope I didn’t break something, at the beginning the renaming worked fine. Maybe it has something to do with the slaves, who knows.

Thanks @Neeloj :+1:t3: