Manually define ClusterID of HTCondor

Question

Does anyone know if it's possible to manually define the starting ClusterID of a new condor installation?

I've wrapped some external services that query condor and they use ClusterID as a reference value.

Unfortunately my instalation got corrupt and the easiest path to solve it is simply to wipe out everything from the /spool directory.

This, however, means that condor will start again with ClusterID=0 which creates havoc on my system.

Answer 1

Answer kindly provided by Ben Cotton from cyclecomputing.com:

SCHEDD_CLUSTER_INITIAL_VALUE is the schedd setting you want.

From the manual:

An integer that specifies the initial cluster number value to use within a job id when a job is first submitted. If the job cluster number reaches the value set by SCHEDD_CLUSTER_MAXIMUM_VALUE and wraps, it will be re-set to the value given by this variable. The default value is 1.

It might be a good idea, if it's possible, to modify your external services to use GlobalJobId to identify jobs instead of just the straight job id. I have a patch submitted (poke, poke, Todd) to add the following to the ClassAd reference:

A globally-unique job identifier. Contains the queue name, JobId, and schedd start time (in seconds since Jan 1, 1970) joined by #. For example, submit.example.com#152.3#1358363336

The inclusion of the schedd start time helps prevent issues like you're encountering (plus eventually the odometer will reset on your cluster ID anyway).