added 2 GPU's and now ethminer aborts

JabulonJabulon Member Posts: 45
Was having no problems on 2 GPU's, so added two more. All devices recognized in aticonfig (and by ethminer). But now after initialising miner, the program aborts with this message:

[xcb] Unknown request in queue while dequeuing
[xcb] Most likely this is a multi-threaded client and XInitThreads has not been called

Can anyone take a stab at what happened here and how to remedy the situation? Do I need to start from scratch and recreate DAG or something? (Ubuntu 14.04.3, 4x R9 280x toxic). Thanks.

Best Answer

Answers

  • JabulonJabulon Member Posts: 45
    additional detail on error message: "ethminer: ../../src/xcb_io.c:179: dequeue_pending_request: Assertion `!xcb_xlib_unknown_req_in_deq' failed.
    Aborted (core dumped)"

    Really mystified, as mining had previously run problem-free and don't I can't sleuth out what might have changed.
  • JabulonJabulon Member Posts: 45
    Yes, I surmised the same after going back to the documentation and noting the 2gb/gpu requirement. Doubled up on RAM. Problem solved. Thanks for the reply.
  • ethfanethfan Member Posts: 458 ✭✭✭
    All my rigs run with 2GB RAM and they are all with 4 or 5 GPUs.
  • MrYukonCMrYukonC Member Posts: 627 ✭✭✭
    ethfan said:

    All my rigs run with 2GB RAM and they are all with 4 or 5 GPUs.

    Yeah, same here and all of my rigs run 6 GPUs.
  • JabulonJabulon Member Posts: 45
    Yeah, it's still a mystery then. When I check in top it is looking like 17.5% of 8gb mem for 2 GPUs, or about 1.4gb mem/GPU. This is consistent across multiple rigs. Ubuntu 14.04/r9 280x. Therefore it's not clear why I'd need to double-up to 16gb RAM for just 4 GPUs, but empirically it seemed to solve the issue I was having with XInitThreads.
  • estevaozabotestevaozabot Member Posts: 29
    You have a problem with crashes?
    My system is fighting with two 280x
  • estevaozabotestevaozabot Member Posts: 29
    another question, you use the power of the risers?
  • JabulonJabulon Member Posts: 45
    The problem vanished after I bumped up the RAM from 8 to 16gb. On that rig I'm now running 4x r9 280x and it's been 100% stable for a week. Not using powered risers. Celeron processor, 1200 watt PSU. Haven't tweaked the clocks or voltages, everything is running at stock. Newly installed catalyst drivers (15.30.1025-x86.x86_64). Power consumption is around 1050 - 1075 watts and I'm getting about 23Mh/s per card. Anyway, that's more info than you asked for. Previously was having no problems running 2 cards, with 8gb Mem. You might try uninstalling/reinstalling your AMD drivers, for lack of other ideas. I will say that once in the past when I had problems with mystery crashes, it turned out to be a bad memory module. If all else fails you might look into that.
  • dont12dont12 Member Posts: 60
    This is weird The Toxic cards are the best, under clock them and they get 27MHS, I have one suggestion its a bit out there but the power supply is too weak for four cards try it on three, with 8Gb of RAM. It may say 1050W but it spikes for small periods of time.
  • JabulonJabulon Member Posts: 45
    I'll try it. What clocks/voltages do you recommend?
  • dont12dont12 Member Posts: 60
    1.2V or slightly over mem 1500 can't remember engine but something like 1000 The toxic cards can push out 27MHS
  • estevaozabotestevaozabot Member Posts: 29
    I am using this extender and am having trouble with crashes ..

    Use it in PCI x16 port motherboard.

    Can you be defective?


  • farwarefarware Member Posts: 116
    Jabulon said:

    The problem vanished after I bumped up the RAM from 8 to 16gb. On that rig I'm now running 4x r9 280x and it's been 100% stable for a week. Not using powered risers. Celeron processor, 1200 watt PSU. Haven't tweaked the clocks or voltages, everything is running at stock. Newly installed catalyst drivers (15.30.1025-x86.x86_64). Power consumption is around 1050 - 1075 watts and I'm getting about 23Mh/s per card. Anyway, that's more info than you asked for. Previously was having no problems running 2 cards, with 8gb Mem. You might try uninstalling/reinstalling your AMD drivers, for lack of other ideas. I will say that once in the past when I had problems with mystery crashes, it turned out to be a bad memory module. If all else fails you might look into that.

    I noticed the exact same thing one some rigs. When I used 2 to 4 GB of RAM the rig would sometimes crash, most often when its recreating the DAG. My conclusion was in certain setups 2GB is not enough to handle this or less likely theres a programming error. Or as you mentioned a faulty module.

    Yet another reason can be power surges .. depends on PSU quality but some cant handle it
Sign In or Register to comment.