Closed ottlukas closed 1 month ago
try my fork: https://github.com/spnettec/plc4x
Anything in that fork, that you would like to contribute back to the project?
We have the same issue for Modbus TCP. Is there something new for a fix of the bug ?
We added a timeout: workerGroup.shutdownGracefully().awaitUninterruptibly(2000);
Which version of PLC4J are you running on? and if possible can you provide a PCAP ?
We use PLC4J 0.11.0. Can't provide a PCAP.
@chrisdutz Does something speak against adding a timeout for the graceful shutdown of the worker groups as suggested by @BioComputerSience ?
No it doesn't ... feel free to whip up a PR.
The PR is open. When will it be released?
Can't find a PR ... would you mind posting a link to this?
But regarding timelines ... I no longer have much time on PLC4X (More on that here https://github.com/chrisdutz/blog/blob/main/plc4x/throwing-the-towel.adoc) ... so I guess there won't be a big release any time soon, the way I currently see it unless someone from the community does it.
Well that is already merged ... so it's in ... you just need to build it yourself.
@chrisdutz As far as I can see the PR was just closed instead of merged. I don't see the relevant changes on "develop" branch.
You are absolutely right ... I wanted to merge it, not sure why I clicked "Close" ... thanks for pointing that out ... and possibly explains why I claimed to have merged it ;-)
@chrisdutz No problem. Thanks for the merge ;-)
I'm polling a modbus device over a RS485 serial connection:
It works for a few minutes and then:
The Migrated 1 channel(s) and Selector.select() log entires continues endless.
I've found that the call in NettyChannelFactory to:
eventExecutors.shutdownGracefully() never returns even though the above log indicates it did. I've changed the call to:
in an attempt to eliminate any possible deadlocking.
Any suggestions would be much appreciated.
Imported from Jira PLC4X-358. Original Jira may contain additional context. Reported by: Kielsholm.