I've started receiving this error on two of the 10 hosts I maintain using NW 8.2 for Linux when the root ( / ) saveset begins. When I observe the backup take place it will backup root as normal until it hits about 11 GB of data then the transfer speed slowly crawls to 0 and the backup hangs. These two hosts have additional mount points (/home,/var, etc) that backup with no problem - this only occurs with the root saveset.
Since this is only causing one mount point to fail I'm not exactly sure how to proceed with troubleshooting. The services have be reset multiple times, server/client reset and /nsr/tmp files have been cleared on both. If I try to backup any files located on root using the save command via the clients cli there's no errors.
Any assistance would be greatly appreciated.
Solved! Go to Solution.
Did you run save from client for root? Try that and if it hangs then do the same again, but -D5 - that should shaw you which file is processing and you get to see where it hangs exactly.
I ran save -D5 / as root and the last few messages remaining after adding the folders is:
client name = clientname.domain.com, client res name = clientname
Skipping setup for 'Backip renamed directories' attribute: level=full, asof=0
Enter setup_saveset_attrs: Save Paths: [/]
lgto_parms for `nsrmmd` successed
got 'store index entries' value of 'Yes'
lgto_auth for `nsrmmd` successed
It has been sitting here for over 20 minutes..
EDIT: I was not specifying my pool using -b - attempting again with D5 to debug.
With debugging enabled I was able to identify a stale mount that was causing Networker to timeout.
We use autofs for NFS mounts and there was a stale autofs mount on these two nodes. When Networker touched this directory it would hang until the timeout was reached.