[Gluster-users] healing but not really, no split-brain - ?

lejeczek peljasz at yahoo.co.uk
Sat Jul 16 12:41:54 UTC 2022


Hi guys

A client node who got shot in the head left some files which 
glusterFS does not seem to be able to heal.
Client had vol mounted off ganesha NFS.
Bits from 'glustershd.log' :
...
[2022-07-16 12:04:47.013045 +0000] E [MSGID: 114031] 
[client-rpc-fops_v2.c:214:client4_0_mkdir_cbk] 
3-USER-HOME-client-1: remote operation failed. 
[{path=(null)}, {errno=22}, {error=Invalid argument}]
...

...
Starting time of crawl: Sat Jul 16 12:41:47 2022

Crawl is in progress
Type of crawl: FULL
No. of entries healed: 0
No. of entries in split-brain: 0
No. of heal failed entries: 21

It's only these specific files. Since then if any healing on 
other files is required then that seems to heal ok.
This is 3-node cluster and all nodes were rebooted just for 
the sake of argument.

And from FS point of view:

-> $ rm -fr /home.ccnr/be1/.local/share/rstudio/pcs
rm: cannot remove '/home.ccnr/be1/.local/share/rstudio/pcs': 
Transport endpoint is not connected

any & all suggestions on how to troubles & fix it are much 
appreciated.
many thanks, L.


More information about the Gluster-users mailing list