operations wrote:From what I understand the rollup that contains the fix for ReFS also contains patches for meltdown and spectre, has anyone tried to extract the patch or updates and just install the ReFS fix so as to avoid the melt/spec patches ?
Gostev wrote:Actually, this is an interesting thought. The pre-release driver was provided by itself, and I know for the fact that it helped every single person who installed it... but now we're installing a huge patch to get the same driver, so it is certainly a possibility something else in this patch is misbehaving. Although this is not very likely, because in that case it would misbehave for everyone. Usually, when experiences are so dramatically different, this indicates some environment-specific trigger such as computer configuration (lack of system resources) or 3rd party software conflict (where antivirus is the primary suspect). I hope those still having issues after installing the patch are able to investigate them with Microsoft.
nmdange wrote:On Windows Server, the mitigations for meltdown and spectre are not enabled by default. You have to create a registry key, and also for spectre it only works if you have updated firmware on your device. Also remember the specter/meltdown patch has been included in the security patches since January, it's not like the ReFS fix is the first time you're getting those patches.
JimmyO wrote:Installed KB4077525 on one of my smaller backup servers with 58TB volume (backup file: 6TB full, 200GB increment) and merge time went from 8h to 1h after first daily backup.
Of course - the server has just been restarted, but first impressions are good
Cicadymn wrote:It looks like this patch had a major affect in getting things to work for me. Despite having 4k block size, I'm no longer crashing and my jobs are completing in a fraction of the time before. I'd see merges that would last 8-12 hours, now they run 20-30 minutes. It's really like a new system! Hopefully things continue to get even better!
Gostev wrote:That's some precious feedback considering there are few people running with 4KB blocks, so we have very little data on such deployments. If you could share, I would be very interested to know the amount of RAM in your backup repository, ReFS volume capacity and utilization, and the largest backup file size. Thanks!