Dear all,
I am starting our PROOF-Cluster setup from scratch. I have some experience with PROOF from the past.
We, at our home institute, have been able to use PROOF but few days ago we just lost our
master machine. We used to have a static master and dynamic workers.
I am working to recover our previous setup. However, I am stuck with some problems.
The local user that starts the xrootd at the master is able to connect to the master:
root [0] TProof *p = TProof::Open(“ayrm@proof.ifca.es:1093”)
Starting master: opening connection …
Starting master: OK
Opening connections to workers: OK (1 workers)
[…]
but local users at the cluster are not able to do so:
root [0] TProof *p = TProof::Open(“arodrig@proof.ifca.es:1093”)
Starting master: opening connection …
Starting master: connection open: setting up server …
| Error condition occured: message from server:
| mst-0:problems accepting callback: process killed - pid: 2043
Info in TXSlave::HandleError: 0x51c7e40:proof.ifca.es:0 got called … fProof: 0x5102c80, fSocket: 0x51c8160 (valid: 1)
Info in TXSlave::HandleError: 0x51c7e40: proof: 0x5102c80
Error in TXSlave::SetupServ: failed to receive slave startup message
TXSlave::HandleError: 0x51c7e40: DONE …
Starting master: failure
PROOF set to sequential mode
The user is known by the master machine.
From the logs in /pool/proofbox I get this error:
121122 16:45:13 2175 xpd-I: ProofServMgr::CreateFork: srvtype = 2
121122 16:45:13 001 Proofx-E: PhyConn::TryConnect: creating UNIX connection to [:0]
121122 16:45:15 001 Proofx-E: PhyConn::TryConnect: creating UNIX connection to [:0]
121122 16:45:17 001 Proofx-E: PhyConn::TryConnect: creating UNIX connection to [:0]
121122 16:45:19 001 Proofx-E: PhyConn::TryConnect: creating UNIX connection to [:0]
121122 16:45:21 001 Proofx-E: PhyConn::TryConnect: creating UNIX connection to [:0]
121122 16:45:23 2175 xpd-E: mst-0: problems accepting callback: process killed - pid: 2208
Do you have any idea of what is going wrong?
Thank you in advance,
Ana.