"How is your SAN client retrieving? Is this by individual files? The recursive needs to walk the filesystem to build its file job queue. This can take a lot of time if you have lots of files. I'm wandering if this is what is happening when you…"
"Before I retrieve , I've checked the fsstate and fsqueue. there is no queue and the drive are free.
Today, I've tested to retrieve again by using the SAN client machine access to data which archive to tape and copy those folder. Then…"
Hi All,I got a problem when I tried to retrieve a huge directory, it took very very long time and then the system seem hang. I'm not sure are there any limitation for retrieve directory? I'm not sure what am I doing right?I've created storage manager policy match with project directory. In each project directory, they took 6-7 LTO-7 cartridges for archive. I have Scalar i6000 with 2xLTO-7 drives work with stornext. When I retrieved the directory, I selected the root directory and it took more…See More
Glad it worked. Sounds like a similar issue. For us the early sn6 had lots of issues with mdarchive. The last one was integer overflow on a couple of as our filesystems. But these have a very high turn over of files.
Thus we now use…"
If you are lucky the issue might be with the MDarchive. We've had similar issues. You could try renaming the "/usr/adic/database/mdarchives/video-mdarchive" to "/usr/adic/database/mdarchives/video-mdarchive.old" and…"
That doesn't look good at all to be honest. Not sure if clearing out the core dumps would help. BTW, the core files should be located under /usr/cvfs/debug.
However, your FS is complaining about "Failed to find object in…"