November 2, 2015 at 7:27 am
Hi,
Had a quick search around and couldn't find someone with the same issue.
Scenario
2 Boxes in HA, identical hardware (VMs) and both utilising identical storage systems (Nimble SANs)
Was needing to do a copy only backup of a few databases for an escrow environment we have, so decided to perform this on the secondary box.
Small ~10GB ish database first, seemed to be really slow, so opened resource monitor and noticed that disk read / write was getting limited to about 15MB/s so took a little while (the big one is over 1TB)
Tried this again on the primary it was done in seconds disk rate in the order of 250MB/s (same settings re compression etc)
Initially had some of the guys looking to see if there were any hardware/networking issues, but whilst they were doing this I decided to make a copy of the backup file on the secondary, the copy disk to disk went in the region of 150MB/s , so it appears that SQL Server is the bottleneck, has anyone else seen this before? and do you have any suggestions?
Thanks,
Andrew
November 2, 2015 at 7:41 am
which hypervisor are you using?
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
November 2, 2015 at 8:10 am
VMWare, doesnt seem to be anything else holding up when looking at processes whilst backup is running as the secondary is currently completely unused apart from as a failover
November 3, 2015 at 3:19 am
one of the it team found the issue, there was a snapshot created of the secondary box a while ago and been left running.
After consolidating it, backup speeds returned to normal
Side effect during the finalisation of the consolidation it dropped connection to the cluster which required the cluster service to be restarted
EDIT : self marked as solution in case its useful to anyone else!
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply