Hi, I'm having trouble with a signal buffer overflow.  It's complicated so please see the details

The buffer overlow is causing our sequences to crash. Instruments are GCMS only (the GCs run fine). It also happens on a colleagues instruments across the building, and it doesn't seem to matter the age of the instrument, or software being used, just that it's an GCMS. The buffer overflow error itself displays in the software and on the GC control panel. The problem is intermittent. Some sequences run fine, others crash on the first sample. Most of the time it overflows between 4-6am after about 6 samples. But not at an exact time like a scheduled task interference. Our IT and our service company both say there's nothing wrong on their end. No IP conflicts, no network outages, nothing physically wrong with the instruments or computer. I'm at a loss. The start of this did occur after a tropical storm caused a power outage, but I wouldn't think the network card would fail *sometimes*.  We didn't install any updates or change configurations recently, and it was working fine previously. I'm thinking of trying a system restore, but would like to avoid it. Any thoughts?

  • Hi gcbill

    The most common cause of this problem is saving GC signals that do not exist. Check under the GC signals to make sure signals are not being saved other than the MSD.



  • Two instruments are running off the one computer (two cards). One is a 6890N/5973N, one is 6890N/5975 with Gerstel Rail. Neither instrument has an FID, just the one MS detector, so only one signal from each. It doesn't seem to matter if both instruments are running. And either one can crash the computer. FYI it had been running fine in this configuration for well over a year.

  • But now you've made me think of something to check...as I understand it, the buffer stores data, and the computer removes it from the buffer (i.e., the GCMS doesn't send the data, just holds it for the computer to take). If for some reason the GC is generating data when it shouldn't be, then the method/computer wouldn't know to retrieve it. It's only expecting MS data. I don't know how that would be possible but I'll check anything at this point. If you have any other ideas please let me know. My service group has already replaced the instrument motherboard, and it didn't solve anything.

  • Hi gcbill, 


    Are you collecting both a GC and MS signal on these systems?    Do you have more than one network card on the PC?  


    Just for clarity, can you tell us what models of GC and MSD you have in your lab?  



  • I just wanted to follow-up to see if this issue is resolved. If so, please come back to the post and click Verify Answer on the best response so it will make the solution more visible.  If you still need help, just let us know and we would be happy to continue working with you

  • James,

    I hope this works, as I'm having trouble getting back in to the community. The problem ended up being a printer. It was printing sometimes and not others, and something about the printer buffer not accepting more information was the GC buffer to become full. All I had to do was remove the printer as the default, and everything worked fine.

    Best Regards

    (Personal details removed by Moderator)

     "james_jenkins" ---04/14/2021 02:08:12 PM---[http://www.agilent.com/] Agilent Community

Was this helpful?