OK,

   an update.  The client I was using was 2.10, so it didn't have the D option 
in its man page.  However I'm not clear on how to update that D status.  The 
underlying OST shows as fine.

some more digging and I was able to find it... I had to look on the OSS not the 
MGS.

w/r,
Kurt

________________________________
From: Kurt Strosahl
Sent: Thursday, June 27, 2019 1:23 PM
To: [email protected]
Subject: lfs df showing a "D" after an ost

Good Afternoon,

    I've got a new 2.12 lustre file system up, and when I run a lfs df it shows 
a D next to one of the OSTs.  I'm not quite sure what has put that D there.  I 
did do a disk replacement test on that OST yesterday, so I'm thinking that it 
is some sort of marker for the underlying zpool being degraded, but the rebuild 
finished more then 24 hours ago.

Looking through the manual (section 13.7) there is a reference to setting OSTs 
to a degraded state, but when I run the get_param it returns an error

lctl get_param obdfilter.*.degraded
error: get_param: param_path 'obdfilter/*/degraded': No such file or directory

lfs df -h
UUID                       bytes        Used   Available Use% Mounted on
lustre19-MDT0000_UUID        2.5T       12.1M        2.5T   0% /lustre19[MDT:0]
lustre19-OST0000_UUID       66.3T        5.4T       60.8T   8% /lustre19[OST:0] 
D
lustre19-OST0001_UUID       66.3T       19.0M       66.3T   0% /lustre19[OST:1]
lustre19-OST0002_UUID       66.3T        4.4T       61.8T   7% /lustre19[OST:2]


w/r,

Kurt J. Strosahl
System Administrator: Lustre, HPC
Scientific Computing Group, Thomas Jefferson National Accelerator Facility
_______________________________________________
lustre-discuss mailing list
[email protected]
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Reply via email to