Rgenoud is Not Completing

[ljbrackney]

Running an rgenoud optimization produces and simulates a number of data points. At some point though, something breaks and SNOW starts barfing:

Error in unserialize(node$con) : error reading from connection
Calls: ... doTryCatch -> recvData -> recvData.SOCKnode -> unserialize
Execution halted
Error in unserialize(node$con) : error reading from connection
Calls: ... doTryCatch -> recvData -> recvData.SOCKnode -> unserialize
Execution halted

Ad infinitum. I've run other algorithms today (LHS, Morris, SPEA, NSGA, PSO) and none of them have trouble completing properly. I also tried redeploying OS Server, but regenoud refuses to complete.

[ljbrackney]

@brianlball determined that this is not related to the algorithm - rather the provisioning method. Things work fine with docker compose but fail when docker swarm is used. This is blocking for both simuwatt and Maalka IN2 projects.

[rHorsey]

Fixed by #213

[ljbrackney]

This apparently has not been resolved. I ran into this last night on 24a.

Error in unserialize(node$con) : error reading from connection
Calls: ... doTryCatch -> recvData -> recvData.SOCKnode -> unserialize
Execution halted

24a will not redeploy properly now.

Perhaps related, but I started an analysis on 24b. It began running okay, and the server suddenly became non-responsive mid-batch. It is presently incommunicado.

[brianlball]

This is a deployment issue, can we close this and make a new issue around deployment?