Occasional Core Error with large slices about orca5 HOT 4 CLOSED

This apparently happened on extend with a large slice:

1   umassvmsite.vm  [ closed, nascent]  
Notices: Reservation 124d66c1-a3e7-4b70-8ffd-df4fe1a82ceb (Slice WAN-80nodes) is in state [Closed,None], err=unexpected ticket failure, message=An error occurred during amend reserve for reservation #42801D70, stack=Exception stack trace: 

orca.shirako.kernel.Kernel.error(Kernel.java:279)
orca.shirako.kernel.Kernel.amendReserve(Kernel.java:95)
orca.shirako.kernel.KernelWrapper.handleReserve(KernelWrapper.java:693)
orca.shirako.kernel.KernelWrapper.ticket(KernelWrapper.java:1009)
orca.shirako.core.ServiceManager.bid(ServiceManager.java:142)
orca.shirako.core.ServiceManager.tickHandler(ServiceManager.java:415)
orca.shirako.core.Actor.actorTick(Actor.java:428)
orca.shirako.core.Actor.access$000(Actor.java:60)
orca.shirako.core.Actor$1.process(Actor.java:333)
orca.shirako.core.Actor.actorMain(Actor.java:377)
orca.shirako.core.Actor$4.run(Actor.java:1018)
java.lang.Thread.run(Thread.java:745)

(80 VMs across multiple sites with a multipoint connection).
It appears the type of reservation is not important - it happened with at least one ION vlan and a bunch of VMs.

from orca5.

Anirban, I wonder if you can look at this.

from orca5.

One more thing - this is apparently not an extend artifact. I think those reservations were never completed - they are all lacking UNIT properties, which suggests that handler never executed. I think they were dead on arrival, which would match Paul's original observations.

from orca5.

This popped out on TAMU (using TAMU SM and broker) where due to a large number of these errors somehow the counts of available resources for VMs and local VLANs went to 0. Errors were on AM/broker container:

2015-09-24 10:35:55,452 [tamu-broker] ERROR orca.tamu-broker - An error occurred during probe pending for reservation #85DC98BB
orca.util.OrcaException: internal error: service overrun in probePending
at orca.shirako.kernel.Reservation.internalError(Reservation.java:739)
at orca.shirako.kernel.BrokerReservation.probePending(BrokerReservation.java:428)
at orca.shirako.kernel.Kernel.probePending(Kernel.java:605)
at orca.shirako.kernel.Kernel.tick(Kernel.java:992)
at orca.shirako.kernel.KernelWrapper.tick(KernelWrapper.java:975)
at orca.shirako.core.Actor.actorTick(Actor.java:433)
at orca.shirako.core.Actor.access$000(Actor.java:60)
at orca.shirako.core.Actor$1.process(Actor.java:333)
at orca.shirako.core.Actor.actorMain(Actor.java:377)
at orca.shirako.core.Actor$4.run(Actor.java:1018)
at java.lang.Thread.run(Unknown Source)
2015-09-24 10:35:55,452 [tamu-broker] ERROR orca.tamu-broker - exception in Kernel.tick
java.lang.RuntimeException: An error occurred during probe pending for reservation #85DC98BB
at orca.shirako.kernel.Kernel.error(Kernel.java:279)
at orca.shirako.kernel.Kernel.probePending(Kernel.java:611)
at orca.shirako.kernel.Kernel.tick(Kernel.java:992)
at orca.shirako.kernel.KernelWrapper.tick(KernelWrapper.java:975)
at orca.shirako.core.Actor.actorTick(Actor.java:433)
at orca.shirako.core.Actor.access$000(Actor.java:60)
at orca.shirako.core.Actor$1.process(Actor.java:333)
at orca.shirako.core.Actor.actorMain(Actor.java:377)
at orca.shirako.core.Actor$4.run(Actor.java:1018)
at java.lang.Thread.run(Unknown Source)
Caused by: orca.util.OrcaException: internal error: service overrun in probePending
at orca.shirako.kernel.Reservation.internalError(Reservation.java:739)
at orca.shirako.kernel.BrokerReservation.probePending(BrokerReservation.java:428)
at orca.shirako.kernel.Kernel.probePending(Kernel.java:605)
... 8 more

from orca5.