[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[AWIPS #NTZ-805469]: Data Disk Completely full!

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [AWIPS #NTZ-805469]: Data Disk Completely full!
  • Date: Fri, 09 Oct 2015 13:33:04 -0600
Hi Bob,

I have experienced something similar on a cloud instance, and discovered this 
week that when the disk was full, each EDEX plugin's purge status was flagged 
in postgres which *prevented* scouring, which is exactly what we do not want to 
happen.  I'm working on a fix and have added a reset option to the edex 
sservice manager (something like "edex purge reset" run automatically every so 
often).

I would check the size of /awips2/edex/data/hdf5/ and manually remove some of 
the larger directories and files which are old.  

You can run this to check for the oldest files in hdf5/

find /awips2/edex/data/hdf5/ -type f -printf '%T+ %p\n' | sort | head -10

See anything extremely old?






> Hi Michael,
> 
> As you know, I recently installed the AWIPS2 software, it
> seemed to be running just fine, but last night the data disk
> filled up:
> 
> # df -h /awips2/
> 
> Filesystem            Size  Used Avail Use% Mounted on
> 
> /dev/mapper/vg_flurry-LogVol06
> 
> 566G  566G   36K 100% /awips2
> 
> 
> There was actually 20K free but then /etc/cron.daily/pypiesLogCleanup.sh
> 
> ran, which freed up an additional 16K.
> 
> 
> As you know this time I installed it on a ZFS file system. I'm not seeing
> the
> 
> error messages about the directory index being full, but nearly all
> the inodes
> 
> are consumed:
> 
> 
> # df -hi /awips2/
> 
> Filesystem           Inodes IUsed IFree IUse% Mounted on
> 
> /dev/mapper/vg_flurry-LogVol06
> 
> 12M   12M   188  100% /awips2
> 
> 
> I can't find it now, but I seem to recall the /awips2 disk was recommend to
> 
> be 500Gb -- as you can see ours is 566Gb, am I remembering incorrectly?
> 
> 
> Is there some data cleanup/garbage collection which is not running
> 
> which will keep the data at a manageable level?
> 
> 
> I notice there are new updates for several components of the AWIPS2
> 
> repository, but I'm sure I don't have space on the disk to install.
> 
> 
> What can I do to clean up the disk? Any other suggestions?
> 
> 
> Thanks.
> 
> 
> -- Bob
> 
> 

Ticket Details
===================
Ticket ID: NTZ-805469
Department: Support AWIPS
Priority: Normal
Status: Open