Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 842. (Read 5805998 times)

That's even better news since it confirms the validity of cgminer's ability to continue doing useful work in the network outage. Thanks.

I can confirm now that at least some of the flurry of submissions were valid.  I can see on my pool stats page that my hashrate dropped while the internet connection was down, but then shot up to 50% higher than my actual hashrate as soon as it returned.  Since they calculate my hashrate based on accepted shares, I believe this means that I send a hugh backlog of valid shares right when the network returned and that made my 5-minute average hashrate look larger than it actually is.

@ewal: Thanks very much for that test.

I will likely be adding a timeout where it can only do locally generated work for 20 minutes since 10 minutes is the average duration it takes for a new block to appear on the network. Best to not keep mining forever with a real network outage.

I tested the latest git code (excluding the very recent kernel tweak) with a simulated internet conncetion failure:

1. I unplugged my internet for 10 minutes and cleared the DNS cache on my miner and on my proxy server.

2. Shortly after the internet connection went down, cgminer reported that the pool was slow and it started doing generating local work.

3. Shortly after that, there was a message about a submit timing out and being cached.

4. Then I waited 10 minutes.  The miner kept mining (at least hashrates on the display stayed high) the entire time.

5. I plugged the internet back in. 

6. There was a flurry of messages that went by very fast.  I assume these were all the cached share submissions.  I didn't see the ones at the start, but the last set were all rejected/stale (which is expected since there was definitly a new block on the network while I was disconnected).

7. After the flurry, cgminer returned to mining as normal without needing to be restarted (and reported it was resuming communications with the server).

So it appears to work better now than when I had my 4 hour internet outage.   At least as far as I can tell in a 10 minute test (and I can't bring myself to test longer than that).

Updated git tree:

Made the kernel code ever so slightly smaller and faster. It does not appear to be significantly faster unless you squint real hard when looking at the hash rates. Anyway, it's there because it's a good thing [tm]. You'll need to do ./autogen.sh again with this change to the tree if building from git.

More stuff planned when time allows.

New windows build:
http://ck.kolivas.org/apps/cgminer/cgminer-1.2.6-win32.zip

You can both read up here and make a judgement:

http://forum.bitcoin.org/index.php?topic=1721.msg350636#msg350636
http://forum.bitcoin.org/index.php?topic=1721.msg353888#msg353888
http://forum.bitcoin.org/index.php?topic=1721.msg358529#msg358529

ATM there's no right or wrong (imo), either way, like the first post said, it works.

Quick bugfix release version 1.2.6:

Source
http://ck.kolivas.org/apps/cgminer/cgminer-1.2.6-1.tar.bz2

(windows build soon hopefully)

- Put a current system status line beneath the total work status line
- Fix a counting error that would prevent cgminer from correctly detecting
situations where getwork was failing - this would cause stalls sometimes
unrecoverably.
- Limit the maximum number of requests that can be put into the queue which
otherwise could get arbitrarily long during a network outage and increase rejects.
- Only count getworks that are real queue requests.
- Allow queued to be zero again since you can now see the queued versus staged value and tune it better. This can also decrease stales.


New status line has the following:
TQ: 2  ST: 2  LS: 0  SS: 0  DW: 0  LW: 0  LO: 0  RF: 0  I: 2

from README:
TQ is Total Queued work items.
ST is STaged work items (ready to use).
LS is Longpoll Staged work items (mandatory new work)
DW is Discarded Work items (work from block no longer valid to work on)
LW is Locally generated Work items (during slow server providing work)
LO is Local generation Occasions (server slow to provide work)
RF is Remote Fail occasions (server slow to accept work)
I is current Intensity (changes in dynamic mode).

If you find the STaged value is often at zero despite your TQ being high then your server may be slow at providing work and you can use this to tune the number of work items in the queue with -Q. It should always be great than 0 if TQ is greater than 0.

This version is slightly faster on average thanks to not slowing down as often during the getworks

Searched out more discussions on other threads and it was decided this was incorrect. I'm leaving it as is with POST after taking it under advisement from other people since I have no actual opinion on the matter.

7 theads per gpu is really not going to help you. Stick with the default of 2. Plus try some other values, like not setting -w or set it to 128 or set intensity to no higher than 8.

Perhaps you did not read that i said CPU miner. For the CPU miner to scan ALL the 4 billion nonces, even 10 minutes are not enough

A kid goes to the doctor and says to the doctor, "Hey Doc, I get a little light headed when I've been hanging upside down from the monkey bars for too long."  The doctor replies, "Then don't hang upside down from the monkey bars for so long."
There's a message here.

Honestly, why would anyone use a scantime of over a minute?  Over 30 seconds doesn't really make a lot of sense in my opinion.  15 seconds is better.  Longpolling is best.  Perhaps an error saying 'Error: User is wasting time computing blocks that were solved 30 seconds ago' might suffice?

Which exactly?

Is that a technical term?

I noticed a bug in the CPU miner which was not in the previous versions. If i set the scantime to very high(above 60) eventually the cpu miner runs out of nonces and shit start to happen.

ckolivas when you get some time please post an windows compiled version too as i don't have mingw on this computer and because of some problems i can't install it now.

Sure! Thanks for your effort.

But still it produces a lot of rejections... 

This is just a reminder

Try -v 2 as well. Since it may pick v 4. But you really should be on -w 128 for a 6850 I think. I lose 5mh/s going from 128 to 256 on a 6870.

I'm disappointed to report,losing like 20mh on 6850. Maybe CGMINER is not configured optimally? This is what i'm using as of the moment. -w 256 -g 7 --intensity 9