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

[LDM #OQE-357383]: (No Subject)

Kevin,

Would you please send me the output of the command

fgrep ldm.pq ~/etc/registry.xml


Also, Stonie Cooper has something to try.

--Steve

---------- Forwarded message ---------
From: Stonie Cooper (he/his/him) <address@hidden>
Date: Fri, Feb 17, 2023 at 6:17 PM
Subject: Re: Impossible ldmadmin(1) problem on Kevin Polston's machine

Steve - I know you sent that to Mike, but I do have a two tests I wouldn't
mind suggesting, if you don't mind.

The first has to do with some idiosyncrasies I have noticed with xfs . . .
in fact, I refuse to use in any longer.  And this is on a modern OS like
Rocky 8 or Ubuntu 22.04.  If Kevin has the ability to grow out a separate
logical volume, and then mount it as /ldmQueue or /data or something like
that.  Make the filesystem ext4, reiserfs, or something other than xfs.
Make "ldm" the owner, and also make sure whatever group he has assigned
"ldm" to:

chown -R ldm:ldm /data

Then edit the registry.xml and point the queue to that fs:

<queue>
    <path>/data/ldm.pq</path>
    <size>500M</size>
    <slots>default</slots>
  </queue>

And that brings me to the second idea . . . have him make his queue much
smaller, and see if that changes anything.

At my former job, I thought I was clever and used xfs as a fs . . . only to
get filesystem "full" notifications when df showed it was only 70%.  Come
to find out, I was out of inodes . . . and when I did a fsck, it wanted to
remove my queue.

In his shoes, that is what I would do to try to eliminate some of the
variables.

>



Ticket Details
===================
Ticket ID: OQE-357383
Department: Support LDM
Priority: Normal
Status: Open
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.