nbd-client /dev/nbd0 is ran twice, causing ldm to segfault

Bug #505787 reported by Alkis Georgopoulos on 2010-01-11
28
This bug affects 5 people
Affects Status Importance Assigned to Milestone
ltsp (Ubuntu)
Undecided
Scott Balneaves

Bug Description

In a fully updated (2010/01/11) Lucid chroot, I get an ldm segfault when I try to enter my username/password.
That's probably because `echo $SERVER` returns "10.160.31.10\n10.160.31.10" (note the enter there).
The SERVER variable is set by ltsp_config, and the problem originates in this line:
  server=$(pgrep -f -l $DEVICE | awk '{print $3}')
In my system, with DEVICE=/dev/nbd0, this pgrep returns two matches:
230 nbd-client 10.160.31.10 2000 /dev/nbd0
232 nbd-client 10.160.31.10 2000 /dev/nbd0

A quick hack would be to include a `head -n 1` there (or use sed/q instead of awk/head), but I don't think two nbd-clients should be there in the first place...

Scott Balneaves (sbalneav) wrote :

ok, I've changed the ltsp_config code to better detect which is the connected nbd. This gets around the double mount in a more robust fashion, and "solves" the problem. Committed upstream.

Scott Balneaves (sbalneav) wrote :

Changed to fixed committed

Changed in ltsp (Ubuntu):
assignee: nobody → Scott Balneaves (sbalneav)
status: New → Fix Committed
tags: added: iso-testing
Scott Balneaves (sbalneav) wrote :

OK, fix released. Should be in lucid updates.

Changed in ltsp (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Duplicates of this bug

Other bug subscribers