Repository navigation
[JENKINS-1364] Hudson should be better in reviving slaves that go down #876
Description
As a user of the master-slave system I think this enhancement is something that
will make hudson practical for scaling up and is really needed thats why I'm
putting it a P1.
Right now the hudson instance sometimes loses slaves - I'm not sure why it
happens, but hey it loses them sometimes. But the thing is I need to manually
launch the slaves even if I've configured them to be started via SSH.
This is really tiresome and hudson should try to revive slaves automatically -
specially if jobs have been queued that require the slave. Having to go there
and manually start slaves is really unacceptable specially when you scale up and
want automation.
Originally reported by akshay_abd, imported from: Hudson should be better in reviving slaves that go down
- status: Closed
- priority: Blocker
- component(s): remoting
- resolution: Fixed
- resolved: 2008-03-06T22:16:22+00:00
- votes: 0
- watchers: 1
- imported: 2025-11-25
Raw content of original issue
As a user of the master-slave system I think this enhancement is something that will make hudson practical for scaling up and is really needed thats why I'm putting it a P1.
Right now the hudson instance sometimes loses slaves - I'm not sure why it happens, but hey it loses them sometimes. But the thing is I need to manually launch the slaves even if I've configured them to be started via SSH.
This is really tiresome and hudson should try to revive slaves automatically - specially if jobs have been queued that require the slave. Having to go there and manually start slaves is really unacceptable specially when you scale up and want automation.
- environment:
Platform: All, OS: All
Activity
jenkins-infra-bot commentedon Mar 6, 2008
akshay_abd:
Raw content of original comment:
Heres the log of the slave going offline, looks like the connection was reset
and the connection lost, and the master never tried to re-establish:
[03/05/08 14:53:27] Launching slave agent
$ ssh mmm.mmm.mmm.com /hudsonslave/startSlave.sh
channel started
Copied maven-agent.jar
Copied maven-interceptor.jar
This is a Unix slave
Read from remote host mmm.mmm.mmm.com: Connection reset by peer
[03/06/08 08:24:31] slave agent was terminated
java.io.EOFException
at
java.io.ObjectInputStream$BlockDataInputStream.peekByte(ObjectInputStream.java:2554)
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1297)
at java.io.ObjectInputStream.readObject(ObjectInputStream.java:351)
at hudson.remoting.Channel$ReaderThread.run(Channel.java:574)
jenkins-infra-bot commentedon Mar 6, 2008
Raw content of original comment:
Implemented in 1.190.
After a drop, Hudson will attempt 6 retries with 5 minutes interval, and after
that it will try every hour.
In my environment, dropped connections are almost unheard of, though.
jenkins-infra-bot commentedon Mar 6, 2008
akshay_abd:
Raw content of original comment:
Sweet
.
Btw can we make the interval configurable? For my setup I would put it every 30
mins, or perhaps less.