Did you uncomment any of the commands in watchdog-cmd.sh?
yeah I did, but running those same commands in other scripts causes the rig to reboot properly.
So what happens now? Do the commands fail when run from that script? Is there any error output? Or does it seem that the script never executes at all?
Everything in that script requires ārootā privileges - are you running the miner as root? I donāt, as a matter of course. What I do is run the miner as my normal user_id, then I start a small script from /etc/rc.local, which is ārootā, that checks every second for a flag file (~/.reboot). If it sees the flag file, which is ātouchedā by watchdog-cmd.sh (the only thing it does), it does the hard reset sequence. What Iāve found is that the hard reset has to be issued immediately after the watchdog script is called in order to work. If thereās too long a delay, like a minute using a cronjob, itās too late to execute the hard reset successfully. YMMV. Happy Holidays!
no error output. Iām assuming the commands fail the computer just freezes and becomes unresponsive, yet it never reboots. I canāt even ping it, just says destination host unreachable.
We have this problem:
[2016-12-18 16:50:32.549] [info] [GPU4] 99.0 I/s 184.6 S/s (5s) 98.2 I/s 182.7 S/s (1m) [2016-12-18 16:50:32.580] [info] [GPU3] 89.0 I/s 169.2 S/s (5s) 89.0 I/s 165.3 S/s (1m) [2016-12-18 16:50:33.027] [error] IO/Error communicating to zec-eu.coinmine.pl [2016-12-18 16:50:33.027] [error] IO/Error communicating to zec-eu.coinmine.pl [2016-12-18 16:50:33.027] [error] IO/Error communicating to zec-eu.coinmine.pl [2016-12-18 16:50:33.048] [info] Connecting to zec-eu.coinmine.pl:8008. [2016-12-18 16:50:33.053] [info] GPU3 Waiting for work from pool. [2016-12-18 16:50:33.054] [info] GPU1 Waiting for work from pool. [2016-12-18 16:50:33.057] [info] GPU0 Waiting for work from pool. [2016-12-18 16:50:33.074] [info] GPU4 Waiting for work from pool. [2016-12-18 16:50:33.090] [info] GPU2 Waiting for work from pool.
I would like to ask when this bug will be eliminated. This usually occurs after reset or rupture Internet. I am away from my rig, and in this case they are idle until a manual reset
I do not think it will be great difficulties to the author, that a miner himself reconnecting with the pool after some downtime. Thanks.
Claymore just posted he his releasing a Linux version of 9.2 with 2% fee
2 Likes
Link?
20 characters .
you want the link of claymore in the optiminer tread loll or OPTIMINER link
On rx 470, I see a 5% hashrate increase with the latest Optiminer over the previous version. Thanks!
i have an average +2 % with new optiminer version
At least a link to claymore stating that he is working on releasing a Linux version would be great. The reason Iām asking is that I couldnāt find claymore stating that he is intending to release a linux version at all.
@coenie really all you had to do was check out his bitcoin talk thread and check out his
latest posts,seems to hard for you so I will lead you by the hand
Re: Claymoreās ZCash AMD GPU Miner v9.2
Today at 02:20:29 PM
Reply with quote #9054
Another news: due to a lot of requests I got, finally I reconsidered my plans about Linux version, I will release it in a couple of days.
changed the GPU card on one Ubuntu 15.10 PC the other day and since that cannot mine either zec or eth. Miner compiles fine, starts, then in 2-3 sec computer freezes completely.
tried to redownload fglrx drivers, etc, nothing works. the rest of PC functions just fine.
one visible problem is that aticonfig have some problems (lists adapters OK, but cannot do odsc commands) and sometimes says that X is not running. I run initialize command, it seem to work, but does not hold on restart.
I know that it is a long shot, but is there any way to revitalize the system, especially that pesky X?
people do have this problem, but i have never seen a solution
Any help is much appreciated.
Appreciate the hand holding, not necessary to be snarky about it, especially since this is on a different forumā¦
I know there have been folks that run into problems on 15.10. Most devs test on 14.04 or 16.04, depending on which driver base they want/need (fglrx, amdgpu-pro). Personally I still find 14.04.3 with fglrx 15.12 drivers to be the most stable and predictable, but then, if you need RX card support, 16.04 is it. I know you can run 14.04.4 with amdgpu-pro drivers, but it takes a bit of work to get it running. The biggest advantage to 16.04/amdgpu-pro is that it does not require X11 to be installed and running. The biggest disadvantage is that you have almost zero GPU control, other than reading temps and setting fans, but even the fan control requires root privileges. Iāve never install 15.10, so I canāt really comment on what issues you may be running into. You might want to check out the ethereum forums for more specific help with that.
That is exactly the reason why I could not post a link to the discussion surely you can understand
thanks, the problem is that this PC has a r7 370 card, I cannot fit Rx 470-480 (maybe short GTX 1060 could fit). therefore, upgrading to 16.04 is not possible as amdgpupro does not support 370.
I can reinstall 14.04 nd start over, but i already install so much stuff (drivers, tweaks, ect) on that PC so it would be kind of wasteful.
Anyway, thanks again.
With Claymore releasing v9.2 for linux64 in the next two days (2% fee), you can put a fork in optiminer, heās well done.
This is true. Optiminer is going to start competing with Claymore, and itās not just about DevFees and hashing power. Claymore has many cool opt-in options.
Competition is always good IMO. Weāll see if Optiminer is up to it.
Iām also eagerly waiting for SilentArmy to get into the ring.
1 Like