Hello
we are in the process of migrating the nodes to CentOS7.
We find that jobs stay IDLE, but there are nodes available in C7.
In the nodes I can see:
==> /var/log/condor/StartLog <==
01/28/19 10:41:43 slot1: Changing activity: Idle -> Benchmarking
01/28/19 10:41:43 BenchMgr:StartBenchmarks()
01/28/19 10:41:46 Initial update sent to collector(s)
01/28/19 10:41:46 Sending DC_SET_READY message to master
<150.244.247.10:9618?addrs=150.244.247.10-9618+[2001-720-420-c003--95]-9618&noUDP&sock=4487_498a>
01/28/19 10:42:09 State change: benchmarks completed
01/28/19 10:42:09 slot1: Changing activity: Benchmarking -> Idle
01/28/19 10:46:46 condor_write(): Socket closed when trying to write
4096 bytes to collector grid003.ft.uam.es, fd is 7
01/28/19 10:46:46 Buf::write(): condor_write() failed
01/28/19 10:51:46 condor_write(): Socket closed when trying to write
4096 bytes to collector grid003.ft.uam.es, fd is 6, errno=104
Connection reset by peer
01/28/19 10:51:46 Buf::write(): condor_write() failed
In the SchedLog of the central node I see that jobs are being rejected:
An excerpt:
01/28/19 11:02:08 (pid:46843) Negotiating for owner:
group_atlas.admin.atl105_score@xxxxxxxxx
01/28/19 11:02:08 (pid:46843) Checking consistency running and
runnable jobs
01/28/19 11:02:08 (pid:46843) Tables are consistent
01/28/19 11:02:08 (pid:46843) Rebuilt prioritized runnable job list in
0.000s.
01/28/19 11:02:08 (pid:46843) Finished negotiating for
group_atlas.admin.atl105_score in local pool: 0 matched, 1 rejected
01/28/19 11:02:08 (pid:46843) Activity on stashed negotiator socket:
<150.244.244.86:22808>
Please, could you help me debugging what could be the issue?