[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] not seeing Windows 7 CPUs



Quoting "Tim St Clair" <tstclair@xxxxxxxxxx>:

Check your START policy, and feel free to send your
NegotiatorLog + StartLog to the list for further inspection.

The condor_config contains:
START=TRUE

After a shutdown and restarting condor_startd and condor_schedd on master, the tail of NegotiatorLog was

--
10/18/12 08:30:13 ---------- Started Negotiation Cycle ----------
10/18/12 08:30:13 Phase 1:  Obtaining ads from collector ...
10/18/12 08:30:13   Getting Scheduler, Submitter and Machine ads ...
10/18/12 08:30:13   Sorting 4 ads ...
10/18/12 08:30:13   Getting startd private ads ...
10/18/12 08:30:13 Got ads: 4 public and 4 private
10/18/12 08:30:13 Public ads include 0 submitter, 4 startd
10/18/12 08:30:13 Phase 2:  Performing accounting ...
10/18/12 08:30:14 Phase 3:  Sorting submitter ads by priority ...
10/18/12 08:30:14 Phase 4.1:  Negotiating with schedds ...
10/18/12 08:30:14  negotiateWithGroup resources used scheddAds length 0
10/18/12 08:30:14 ---------- Finished Negotiation Cycle ----------
10/18/12 08:31:14 ---------- Started Negotiation Cycle ----------
10/18/12 08:31:14 Phase 1:  Obtaining ads from collector ...
10/18/12 08:31:14   Getting Scheduler, Submitter and Machine ads ...
10/18/12 08:31:14   Sorting 4 ads ...
10/18/12 08:31:14   Getting startd private ads ...
10/18/12 08:31:14 Got ads: 4 public and 4 private
10/18/12 08:31:14 Public ads include 0 submitter, 4 startd
10/18/12 08:31:14 Phase 2:  Performing accounting ...
10/18/12 08:31:14 Phase 3:  Sorting submitter ads by priority ...
10/18/12 08:31:14 Phase 4.1:  Negotiating with schedds ...
10/18/12 08:31:14  negotiateWithGroup resources used scheddAds length 0
10/18/12 08:31:14 ---------- Finished Negotiation Cycle ----------
--

StartLog tail was

--
10/18/12 08:32:05 Locale: English_United States.1252
10/18/12 08:32:05 Setting maximum accepts per cycle 8.
10/18/12 08:32:05 ******************************************************
10/18/12 08:32:05 ** condor_startd (CONDOR_STARTD) STARTING UP
10/18/12 08:32:05 ** C:\Programs\condor\bin\condor_startd.exe
10/18/12 08:32:05 ** SubsystemInfo: name=STARTD type=STARTD(7) class=DAEMON(1)
10/18/12 08:32:05 ** Configuration: subsystem:STARTD local:<NONE> class:DAEMON
10/18/12 08:32:05 ** $CondorVersion: 7.8.4 Sep 18 2012 BuildID: 64675 $
10/18/12 08:32:05 ** $CondorPlatform: x86_64_winnt_6.1 $
10/18/12 08:32:05 ** PID = 3292
10/18/12 08:32:05 ** Log last touched 10/18 07:22:49
10/18/12 08:32:05 ******************************************************
10/18/12 08:32:05 Using config source: C:\programs\condor\condor_config
10/18/12 08:32:05 Using local config sources:
10/18/12 08:32:05    C:\programs\condor/condor_config.local
10/18/12 08:32:05 DaemonCore: command socket at <<IP>:50961>
10/18/12 08:32:05 DaemonCore: private command socket at <<IP>:50961>
10/18/12 08:32:05 Setting maximum accepts per cycle 8.
10/18/12 08:32:06 VM-gahp server reported an internal error
10/18/12 08:32:06 VM universe will be tested to check if it is available
10/18/12 08:32:06 History file rotation is enabled.
10/18/12 08:32:06   Maximum history file size is: 20971520 bytes
10/18/12 08:32:06   Number of rotated history files is: 2
10/18/12 08:32:06 slot1: New machine resource allocated
10/18/12 08:32:06 slot2: New machine resource allocated
10/18/12 08:32:06 slot3: New machine resource allocated
10/18/12 08:32:06 slot4: New machine resource allocated
10/18/12 08:32:06 slot5: New machine resource allocated
10/18/12 08:32:06 slot6: New machine resource allocated
10/18/12 08:32:06 slot7: New machine resource allocated
10/18/12 08:32:06 slot8: New machine resource allocated
10/18/12 08:32:11 CronJobList: Adding job 'mips'
10/18/12 08:32:11 CronJobList: Adding job 'kflops'
10/18/12 08:32:11 CronJob: Initializing job 'mips' (C:\programs\condor/bin/condor_mips.exe) 10/18/12 08:32:11 CronJob: Initializing job 'kflops' (C:\programs\condor/bin/condor_kflops.exe)
--

and MasterLog tail was:

--
10/18/12 08:28:10 Locale: English_United States.1252
10/18/12 08:28:10 Setting maximum accepts per cycle 8.
10/18/12 08:28:10 ******************************************************
10/18/12 08:28:10 ** condor (CONDOR_MASTER) STARTING UP
10/18/12 08:28:10 ** C:\programs\condor\bin\condor_master.exe
10/18/12 08:28:10 ** SubsystemInfo: name=MASTER type=MASTER(2) class=DAEMON(1)
10/18/12 08:28:10 ** Configuration: subsystem:MASTER local:<NONE> class:DAEMON
10/18/12 08:28:10 ** $CondorVersion: 7.8.4 Sep 18 2012 BuildID: 64675 $
10/18/12 08:28:10 ** $CondorPlatform: x86_64_winnt_6.1 $
10/18/12 08:28:10 ** PID = 2192
10/18/12 08:28:10 ** Log last touched 10/18 07:22:55
10/18/12 08:28:10 ******************************************************
10/18/12 08:28:10 Using config source: C:\programs\condor\condor_config
10/18/12 08:28:10 Using local config sources:
10/18/12 08:28:10    C:\programs\condor/condor_config.local
10/18/12 08:28:10 DaemonCore: command socket at <<IP>:50836>
10/18/12 08:28:10 DaemonCore: private command socket at <<IP>:50836>
10/18/12 08:28:10 Setting maximum accepts per cycle 8.
10/18/12 08:28:10 Authorized application C:\programs\condor/bin/condor_negotiator.exe is now enabled in the firewall. 10/18/12 08:28:10 Authorized application C:\programs\condor/bin/condor_collector.exe is now enabled in the firewall. 10/18/12 08:28:11 Authorized application C:\programs\condor/bin/condor_starter.exe is now enabled in the firewall. 10/18/12 08:28:11 Authorized application C:\programs\condor/bin/condor_vm-gahp.exe is now enabled in the firewall. 10/18/12 08:28:11 Authorized application C:\programs\condor/bin\condor_dagman.exe is now enabled in the firewall. 10/18/12 08:28:11 Started DaemonCore process "C:\programs\condor/bin/condor_collector.exe", pid and pgroup = 948 10/18/12 08:28:11 Waiting for C:\programs\condor/log/.collector_address to appear.
10/18/12 08:28:12 Found C:\programs\condor/log/.collector_address.
10/18/12 08:28:13 Started DaemonCore process "C:\programs\condor/bin/condor_negotiator.exe", pid and pgroup = 2056
--

The only difference between "condor_status" and "condor_status -debug" is the first line shows the locale information. The jobs I have submitted to the three Windows XP machines were using the vanilla universe.

The Windows firewall is disabled on all the machines, but all use "Microsoft Forefront Endpoint Protection". I think FEP turns off the firewall, but the settings should be the same across the "cluster".