[Gluster-users] Failed to populate loc for thin-arbiter

[lejeczek]

On 10/03/2021 04:10, Ravishankar N wrote:
>
> On 09/03/21 11:43 pm, lejeczek wrote:
>> Hi guys,
>>
>> I have a simple volume but which seems to suffer from 
>> some problems. (maybe all volumes in the cluster also)
>>
>> ...
>> [2021-03-09 17:59:08.195634] E [MSGID: 114058] 
>> [client-handshake.c:1455:client_query_portmap_cbk] 
>> 0-USER-HOME-ta-2: failed to get the port number for 
>> remote subvolume. Please run 'gluster volume status' on 
>> server to see if brick process is running.
>> [2021-03-09 17:59:18.192257] E [MSGID: 108044] 
>> [afr-common.c:3067:afr_ta_id_file_check] 
>> 0-USER-HOME-replicate-0: Failed to lookup/create 
>> thin-arbiter id file. [Transport endpoint is not connected]
>
> It appears as if your fuse mount is not able to connect to 
> the thin arbiter node. Is the glusterfsd process running 
> on the thin-arbiter node?
>
> Ideally, the ID file is created by the first client (fuse 
> mount) that mounts and accesses a newly created thin 
> arbiter volume for the first time.
>
> Regards,
>
> Ravi
>
Yes it seems even more puzzling to me. I can see that volume 
with 'arbiter' also misbehave while I though problem was 
only with 'thin-arbiter'
I do not suppose that have anything to do with 'samba' which 
runs 'shares' off glusterfs volumes?
On all the nodes involved in a volume I see in 'glustershd.log'
...

[2021-03-10 09:26:24.500574] I 
[rpc-clnt.c:1971:rpc_clnt_reconfig] 13-USER-HOME-client-1: 
changing port to 49157 (from 0)
[2021-03-10 09:26:27.506918] I 
[rpc-clnt.c:1971:rpc_clnt_reconfig] 13-USER-HOME-client-1: 
changing port to 49157 (from 0)
[2021-03-10 09:26:30.513109] I 
[rpc-clnt.c:1971:rpc_clnt_reconfig] 13-USER-HOME-client-1: 
changing port to 49157 (from 0)
[2021-03-10 09:26:33.521548] I 
[rpc-clnt.c:1971:rpc_clnt_reconfig] 13-USER-HOME-client-1: 
changing port to 49157 (from 0)
[2021-03-10 09:29:10.487079] I 
[socket.c:865:__socket_shutdown] 2-DATA-ta-2: intentional 
socket shutdown(5)
[2021-03-10 09:29:10.490647] I 
[socket.c:865:__socket_shutdown] 4-WORK-ta-2: intentional 
socket shutdown(5)
...

That is a freshly re/created volumes. 'peer status' shows no 
issues.
In terms of 'networking' also I do find any issues, only 
thing I wonder of is - do I, or rather glusterfs, must have 
hostnames in DNS or 'etc/hosts' only is perfectly fine.

Again, suggestions on how to troubleshoot this are greatly 
appreciated.
thanks, L.


> [2021-03-09 17:59:29.125268] E [MSGID: 108044] 
> [afr-transaction.c:1151:afr_ta_post_op_do] 
> 0-USER-HOME-replicate-0: Failed to populate loc for 
> thin-arbiter. [Invalid argument]
>> [2021-03-09 17:59:29.125366] E [MSGID: 108044] 
>> [afr-transaction.c:763:afr_changelog_post_op_fail] 
>> 0-USER-HOME-replicate-0: Failing CREATE for gfid 
>> 00000000-0000-0000-0000-000000000000. Fop state is:1 
>> [Invalid argument]
>> [2021-03-09 17:59:29.235173] E [MSGID: 108006] 
>> [afr-common.c:5309:__afr_handle_child_down_event] 
>> 0-USER-HOME-replicate-0: All subvolumes are down. Going 
>> offline until at least one of them comes back up.
>> [2021-03-09 18:02:39.737104] E [MSGID: 114058] 
>> [client-handshake.c:1455:client_query_portmap_cbk] 
>> 0-USER-HOME-ta-2: failed to get the port number for 
>> remote subvolume. Please run 'gluster volume status' on 
>> server to see if brick process is running.
>> ...
>>
>> Any suggestions as to what can be the problem much 
>> appreciated.
>> many thanks, L.
>> ________
>>
>>
>>
>> Community Meeting Calendar:
>>
>> Schedule -
>> Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC
>> Bridge: https://meet.google.com/cpu-eiue-hvk
>> Gluster-users mailing list
>> Gluster-users at gluster.org
>> https://lists.gluster.org/mailman/listinfo/gluster-users
>>
>