HTTPCMD Hangs Up

[JeffS]

I have several BRX PLCs across the country sending data to a remote server.  The problem I have is that one of them stopped sending data, but after remoting into the device I see that it is trying to send data and acting like the server is never responding, so it gives operation timed out as the error.  I have seen this happen in the past and a power cycle fixes it.  What I wanted to know is if there is anything I can pull from this device to help determine why this is timing out in the first place since the server is available and working with other BRX PLCs setup the same way.

The device is currently timing out and will continue to do so until I have the PLC power cycled.

[BobO]

I have several BRX PLCs across the country sending data to a remote server.  The problem I have is that one of them stopped sending data, but after remoting into the device I see that it is trying to send data and acting like the server is never responding, so it gives operation timed out as the error.  I have seen this happen in the past and a power cycle fixes it.  What I wanted to know is if there is anything I can pull from this device to help determine why this is timing out in the first place since the server is available and working with other BRX PLCs setup the same way.

The device is currently timing out and will continue to do so until I have the PLC power cycled.

Not really, at least not with production firmware. If there is some kind of recurrent failure that can be duplicated, it would be best for us to try to dupe it here. Barring that, we can build some custom code that allows us some visibility into internal driver states, which can be very helpful when stuff hangs.

[JeffS]

Sorry, I don't know what exactly causes this to duplicate it.  All I know is the PLC continues to operate as normal other than the HTTPCMD stuff.  Still communicates to other BRX and DL PLCs as well as HMIs via ethernet.  I can remote into the PLC fine over ethernet. 

The only way I typically notice when this happens is when a customer complains about a report missing data, since the server side just collects data and can't reconcile a stop in production vs a stop in communications.  It has only happened a few times now, and a power cycle fixes it, just wanted to try and find out why it is happening so we don't keep having this problem crop up.

[belias]

Hello,

Just wanted to mention that I've seen something similar and posted about it here:

https://forum.hosteng.com/index.php?topic=3330.msg25461#msg25461

BobO was great about taking the DMD file and attempting to replicate the problem, but unfortunately he wasn't able to make it happen.  In my case, it happens about once every 30-60 days, and it also requires a full power cycle to correct.

I just had another issue happen (which I why I logged onto the forum) regarding the MQTT broker hanging...I'm going to make a different post about that one.


- Brian

[BobO]

We found an issue between the MAC and PHY in the Ethernet, and we think we have it fixed. That particular problem takes all Ethernet down, not just single protocols, so that may be a different issue.

[belias]

Hmm...OK.  In my case with the HTTPCMD issue, I was still able to access PLC via Do-More by ethernet.  Other ethernet functions worked also (MQTT, NTP, etc.).  So it may indeed be a different issue.  Thanks for the update though.