OpenVZ Forum


Home » General » Support » [SOLVED] Online migration always fails
[SOLVED] Online migration always fails [message #35982] Sat, 09 May 2009 18:37 Go to next message
divB is currently offline  divB
Messages: 79
Registered: April 2009
Member
Hi,

Online migration works fine with a few VEs but I have three VEs which always fail while live-migration Sad

# vzmigrate --remove-area no --keep-dst --online -v 192.168.200.2 201
OPT:--remove-area
OPT:--keep-dst
OPT:--online
OPT:-v
OPT:192.168.200.2
StartingOpenVZ is running...
201.conf                                                                                                                      100% 1756     1.7KB/s   00:00
Saved parameters for VE 201
         SyncingLive   Setting up checkpoint...
        suspend...
        get context...
Checkpointing completed succesfully
   Setting up checkpoint...
        join context..
        dump...
Can not dump VE: Device or resource busy
Error: splice is used by 24743,907(apache2)
Checkpointing failed
Error: Resuming...


What is the problem here?

Thank you,
divB

[Updated on: Sat, 09 May 2009 22:03]

Report message to a moderator

Re: Online migration always fails [message #35983 is a reply to message #35982] Sat, 09 May 2009 18:53 Go to previous message
divB is currently offline  divB
Messages: 79
Registered: April 2009
Member
Sorry I have found the Bug:

* http://bugzilla.openvz.org/show_bug.cgi?id=881

Is there any workaround?

divB

/EDIT: Not the elegant way, but I resolved it this way:

--- /usr/sbin/vzmigrate 2009-07-06 16:41:51.000000000 +0200
+++ /usr/local/sbin/vzmigrate2  2009-07-08 01:17:45.000000000 +0200
@@ -446,6 +446,9 @@
 if [ $online -eq 1 ]; then
        log 1 "Live migrating VE"

+       log 1 "Stop apache2 if it is installed"
+       vzctl exec $VEID '[ -f /etc/init.d/apache2 ] && /etc/init.d/apache2 stop'
+
        log 2 "Suspending VE"
        time_suspend=$(date +%s.%N)
        if ! logexec 2 vzctl chkpnt $VEID --suspend ; then
@@ -560,6 +563,9 @@
        log 2 "Removing dumpfiles"
        rm -f "$VE_DUMPFILE"
        $SSH "root@$host" "rm -f $VE_DUMPFILE"
+
+       log 1 "Start apache2 again if it is installed"
+       $SSH root@$host vzctl exec $VEID "'[ -f /etc/init.d/apache2 ] && /etc/init.d/apache2 start'"
 else
        if [ "$state" = "running" ]; then
                log 1 "Starting VE"


divB

[Updated on: Sat, 09 May 2009 22:02]

Report message to a moderator

Previous Topic: Deleted libraries in debian VE
Next Topic: [SOLVED] Errors when resuming live-migration
Goto Forum:
  


Current Time: Thu Oct 17 10:08:52 GMT 2024

Total time taken to generate the page: 0.05364 seconds