Since upgrading our condor central managers on to 8.2.1 (running on Scientific Linux release 6.5) we have been experiencing matching problems to workers running condor 8.0.6. For example:
The Requirements _expression_ for your job reduces to these conditions:
    ÂSlots
Step  ÂMatched ÂCondition
----- Â-------- Â---------
[0] Â Â Â Â Â 4 ÂVMType is "atlas-worker"
[1] Â Â Â Â Â 4 ÂTarget.Arch == "x86_64"
[3] Â Â Â Â Â 4 ÂTARGET.OpSys == "LINUX"
[5] Â Â Â Â Â 4 ÂTARGET.Disk >= RequestDisk
[7] Â Â Â Â Â 4 ÂTARGET.Memory >= RequestMemory
[9] Â Â Â Â Â 4 ÂTARGET.HasFileTransfer
Suggestions:
  Condition             Machines Matched  ÂSuggestion
  ---------             ----------------  Â----------
1 Â VMType is "atlas-worker" Â Â Â Â Â0 Â Â Â Â Â Â Â Â Â MODIFY TO "atlas-worker"
2 Â Target.Arch == "x86_64" Â Â Â Â Â 4 Â Â Â Â Â Â Â Â Â Â
3 Â ( TARGET.OpSys == "LINUX" ) Â Â Â 4 Â Â Â Â Â Â Â Â Â Â
4 Â ( TARGET.Disk >= 10000000 ) Â Â Â 4 Â Â Â Â Â Â Â Â Â Â
5 Â ( TARGET.Memory >= 29500 ) Â Â Â Â4 Â Â Â Â Â Â Â Â Â Â
6 Â ( TARGET.HasFileTransfer ) Â Â Â Â4Â
These two strings ("atlas-worker") look the same, yet the match fails. I've set the logging options to full debug but the only output I see in the MatchLog is that the match fails. With the same setup running on Scientific Linux release 5.10 works. Has anyone seen similar problems? Any idea what may be wrong?