I would recommend you to check the brick logs , then the gluster logs and last the vdsm
log.
At least vdsm should timeout if it can't create the task in a reasonable time frame
... or maybe not ?
Best Regards,Strahil Nikolov
В сряда, 24 април 2019 г., 19:50:51 ч. Гринуич-4, Steffen Luitz
<ovirt(a)carlui.net> написа:
This is on a 3-node hyperconverged environment with glusterfs.
After upgrading to oVirt 4.3.3 (from 4.3.2) creating a new disk takes very long (hours for
a 100GByte disk, making it essentially impossible to create a new disk image).
In the UI the default is "preallocated" but changing it to thin provision does
not make any difference. Regardless of this setting, an fallocate process gets started on
the SDM host:
/usr/bin/python2 /usr/libexec/vdsm/fallocate 107374182400
/rhev/data-center/mnt/glusterSD/s-vmhost2-ovir.t...
Using fallocate to create a file directly on the underlying file system is fast. Using it
to create a file through the glusterfs fuse mount is very slow.
Thanks for any insights.
Steffen
_______________________________________________
Users mailing list -- users(a)ovirt.org
To unsubscribe send an email to users-leave(a)ovirt.org
Privacy Statement:
https://www.ovirt.org/site/privacy-policy/
oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives:
https://lists.ovirt.org/archives/list/users@ovirt.org/message/AMDKSAFEUCU...