On 01/06/17 19:26, John White wrote: > Good Morning, > I’ve got a robinhood node with 3 file systems being monitored. 2 are > working just fine. Initial scan + continuing changelog monitors running > without incident, incrementing and decrementing as expected. The 3rd will > scan just fine but it’s simply not taking in any changelogs. Upon starting > the daemon, I do see the following (but I see the same for the others, so I’m > not sure how pertinent it is: > > 2017/01/06 10:21:36 [18299/1] CheckFS | '/clusterfs/etna' matches mount point > '/clusterfs/etna', type=lustre, fs=10.4.205.50@o2ib:/vulcan2 > 2017/01/06 10:21:36 [18299/1] llapi | warning: llapi_changelog_start() called > w/o CHANGELOG_FLAG_JOBID > 2017/01/06 10:21:36 [18299/1] Main | Daemon started (running modules: > log_reader) > > Further, even while running an active IO workload, I don’t really see any > changes with the ‘lfs changelog’ command. Any ideas? Hello,
It is normal lfs changelog does not return records if robinhood consumed them. In its log file, robinhood regularly reports statistics about the changelogs it has read (grep STATS in it). This will indicate you if it has read records. Else, check a changelog reader is properly registered on the MDT and that the changelog mask is not empty. Thomas ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ robinhood-support mailing list robinhood-support@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/robinhood-support
