You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@cloudstack.apache.org by "Wei Zhou (JIRA)" <ji...@apache.org> on 2013/07/31 15:27:50 UTC
[jira] [Commented] (CLOUDSTACK-3947) haproxy is running but
haproxy.pid file is missing in virtual router
[ https://issues.apache.org/jira/browse/CLOUDSTACK-3947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13725222#comment-13725222 ]
Wei Zhou commented on CLOUDSTACK-3947:
--------------------------------------
Thanks my colleagues Shashi Dahal and Mukesh Kumar reproducing this issue.
There is a sentence "sleep 2" in /root/reconfigLB.sh in the virtual routers ( see below)
If a command pause at "sleep 2" while another command come in, this issue will appear.
CloudStack has the mechanism to prevent concurrent operations on the same network/ipaddress/vpc by overriding getSyncObjType and getSyncObjId methods.
However, these two methods is not overrided in CreateLBStickinessPolicyCmd.
I will commit a patch for this issue. I will fix similar issues later.
------------------------------------------/root/reconfigLB.sh in VR------------------------------------------
ret=0
# save previous state
mv /etc/haproxy/haproxy.cfg /etc/haproxy/haproxy.cfg.old
mv /var/run/haproxy.pid /var/run/haproxy.pid.old
mv /etc/haproxy/haproxy.cfg.new /etc/haproxy/haproxy.cfg
kill -TTOU $(cat /var/run/haproxy.pid.old)
sleep 2
if haproxy -D -p /var/run/haproxy.pid -f /etc/haproxy/haproxy.cfg; then
logger -t cloud "New haproxy instance successfully loaded, stopping previous one."
kill -KILL $(cat /var/run/haproxy.pid.old)
rm -f /var/run/haproxy.pid.old
ret=0
else
logger -t cloud "New instance failed to start, resuming previous one."
kill -TTIN $(cat /var/run/haproxy.pid.old)
rm -f /var/run/haproxy.pid
mv /var/run/haproxy.pid.old /var/run/haproxy.pid
mv /etc/haproxy/haproxy.cfg /etc/haproxy/haproxy.cfg.new
mv /etc/haproxy/haproxy.cfg.old /etc/haproxy/haproxy.cfg
ret=1
fi
exit $ret
------------------------------------------------------------------------------------
> haproxy is running but haproxy.pid file is missing in virtual router
> --------------------------------------------------------------------
>
> Key: CLOUDSTACK-3947
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-3947
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the default.)
> Affects Versions: 4.0.2
> Environment: cloudstack 4.0.2
> Reporter: Wei Zhou
> Assignee: Wei Zhou
> Priority: Critical
>
> I think this issue exists on 4.1/4.2 as well.
> agent.logs
> 2013-07-30 11:01:01,546 DEBUG [resource.virtualnetwork.VirtualRoutingResource] (agentRequest-Handler-5:null) + iflag=+ aflag=+ dflag=+ fflag=+ sflag=+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ iflag=1+ domRIp=169.254.1.253+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ fflag=1+ cfgfile=/tmp/169_254_1_2537137377844466896245cfg+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ aflag=1+ addedIps=10.11.102.174:111:,+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ sflag=1+ statsIps=10.11.102.174:8081:0/0:,,+ getopts i:a:d:f:s: OPTION+ cert=/root/.ssh/id_rsa.cloud+ '[' 11 '!=' 11 ']'+ check_gw 169.254.1.253+ ping -c 1 -n -q 169.254.1.253+ '[' 0 -gt 0 ']'+ return 0+ '[' 0 -gt 0 ']'+ copy_haproxy 169.254.1.253 /tmp/169_254_1_2537137377844466896245cfg+ local domRIp=169.254.1.253+ local cfg=/tmp/169_254_1_2537137377844466896245cfg+ scp -P 3922 -q -o StrictHostKeyChecking=no -i /root/.ssh/id_rsa.cloud /tmp/169_254_1_2537137377844466896245cfg root@169.254.1.253:/etc/haproxy/haproxy.cfg.new+ return 0+ '[' 0 -gt 0 ']'+ ssh -p 3922 -q -o StrictHostKeyChecking=no -i /root/.ssh/id_rsa.cloud root@169.254.1.253 '/root/loadbalancer.sh -i 169.254.1.253 -f /tmp/169_254_1_2537137377844466896245cfg -a 10.11.102.174:111:, -s 10.11.102.174:8081:0/0:,,'mv: cannot stat `/var/run/haproxy.pid': No such file or directory[ALERT] 210/090100 (9007) : Starting proxy stats_on_public: cannot bind socket[ALERT] 210/090100 (9007) : Starting proxy 10_11_102_174-111: cannot bind socketcat: /var/run/haproxy.pid.old: No such file or directorykill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]mv: cannot stat `/var/run/haproxy.pid.old': No such file or directory+ exit 1
> 2013-07-30 11:01:01,546 DEBUG [cloud.agent.Agent] (agentRequest-Handler-5:null) Seq 6-1836847202: { Ans: , MgmtId: 345051509349, via: 6, Ver: v1, Flags: 0, [{"Answer":{"result":false,"details":"+ iflag=+ aflag=+ dflag=+ fflag=+ sflag=+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ iflag=1+ domRIp=169.254.1.253+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ fflag=1+ cfgfile=/tmp/169_254_1_2537137377844466896245cfg+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ aflag=1+ addedIps=10.11.102.174:111:,+ getopts i:a:d:f:s: OPTION+ case $OPTION in+ sflag=1+ statsIps=10.11.102.174:8081:0/0:,,+ getopts i:a:d:f:s: OPTION+ cert=/root/.ssh/id_rsa.cloud+ '[' 11 '!=' 11 ']'+ check_gw 169.254.1.253+ ping -c 1 -n -q 169.254.1.253+ '[' 0 -gt 0 ']'+ return 0+ '[' 0 -gt 0 ']'+ copy_haproxy 169.254.1.253 /tmp/169_254_1_2537137377844466896245cfg+ local domRIp=169.254.1.253+ local cfg=/tmp/169_254_1_2537137377844466896245cfg+ scp -P 3922 -q -o StrictHostKeyChecking=no -i /root/.ssh/id_rsa.cloud /tmp/169_254_1_2537137377844466896245cfg root@169.254.1.253:/etc/haproxy/haproxy.cfg.new+ return 0+ '[' 0 -gt 0 ']'+ ssh -p 3922 -q -o StrictHostKeyChecking=no -i /root/.ssh/id_rsa.cloud root@169.254.1.253 '/root/loadbalancer.sh -i 169.254.1.253 -f /tmp/169_254_1_2537137377844466896245cfg -a 10.11.102.174:111:, -s 10.11.102.174:8081:0/0:,,'mv: cannot stat `/var/run/haproxy.pid': No such file or directory[ALERT] 210/090100 (9007) : Starting proxy stats_on_public: cannot bind socket[ALERT] 210/090100 (9007) : Starting proxy 10_11_102_174-111: cannot bind socketcat: /var/run/haproxy.pid.old: No such file or directorykill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]mv: cannot stat `/var/run/haproxy.pid.old': No such file or directory+ exit 1","wait":0}}] }
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira