JIRA can't create a lock for your jira.home directory
cvandome opened this issue · 13 comments
Hello all,
I've a wired issue, my Jira setup mount a nfs volume for Jira Home, on the first start, files are written to NFS share with the right user (daemon or uid 1 on server side), the .lock_file is also written but Jira show me this error.
I don't understand, if someone could help me.
Here is the NFS Export line from the server:
/nfs/exports/jira-home 192.168.205.0/24(rw,async,no_subtree_check,no_root_squash)
Thanks !
Don't know anything about NFS sorry, but what error does JIRA show?
Thanks for reply, Jira show on the webui: JIRA can't create a lock for your jira.home directory
It's wired because when I look inside the NFS folder, I see that Jira can write in this folder.
I've already tested this but no success :(
I'll continue to inspect and come back to post answer here, it can helps.
Best regards
Cédric Vandôme
When I update docker image and restart ,I got "Setup: JIRA couldn't create the jira.home directory“ issue. try find the "lock" file to delete,I faild. where can I find it. can anyone help me? it drive me crazy .
Same issue
Problem solved by removing the symlink of indexes dir and re-put it in nfs in its own place , accepting the luncene warnings in jira to use nfs for indexes.
unlink /nfs/web-hx/coresrv/projects/jira/dev/jira_home/caches/indexes
Can also be solved by running the container as previlged
hello @iahmad-khan I've fixed issue with NFS config also, it's a NFS side issue
For help it works for me:
On every NFS client:
yum install nfs-utils nfs-utils-lib
systemctl enable rpcbind
systemctl enable nfs-lock
systemctl enable nfs-idmap
systemctl start rpcbind
systemctl start nfs-lock
systemctl start nfs-idmap
In each /etc/fstab client:
192.168.X.X:/nfs/exports /data nfs rw,sync,hard,intr 0 0
We can close this issue
hi ,
I have already these packages installed and mount configured , but it stops working when i create a symlink for indexs directry to local file system , and permissions are granted to the user on that local directory where indexes are pointing to from nfs.
so this problem occurs when we symlink indexs dir to local file system and keep the home dir on nfs
sorry but each NFS probleme is very specific and is out of scope of this issue.
Good luck !
If you're using latest
the UID has changed caused by moving to Alpine as base image