BITMAIN AntRouter R1: 1st wireless networking device with bitcoin mining chip

[aarons6]

Yes it is actually hashing as reported by Antpool. That said I have not looked too closely just restarted it. The slight complication is that I run at 15GH & usually when it "goes slow" it is showing the standard speed of 5GH, so although I do not think this is the case, I had wondered if it had somehow picked up the default frequency from somewhere?

Will take a closer look next time it does this. Is there anything else I can look at to help with the problem?


Rich

you can ssh in and run cgminer-api to see what it says.. it would give you more of a direct stats then the ui or the pool..

if it shows the 5s at 15ghs but the avg at 5gsh then its just averaging wrong and everything is ok. you can restart it but its prob not needed.

i can show you a quick script that will restart if gsh goes low..

put this in a text and put it in /usr/bin and make it 0755 then run it with scheduled tasks running every 5 or 10 minutes.

###START
minhash=1400
SPEED=`cgminer-api | grep 'GHS av] => ' | cut -c 16- | tr -d '.'`

  if [ "$SPEED" -lt $minhash ]; then
    logger "cgminer was restarted for low hash"
    /etc/init.d/cgminer stop
    killall -9 cgminer
    sleep 1
    /etc/init.d/cgminer start
    exit 0;

fi
###END

what this will do is check the avg ghs, if its less then 14 it will restart.. you can change the min to whatever. no decimal.

[1715533164]

1715533164

[1715533164]

1715533164

[1715533164]

1715533164

[1715533164]

1715533164

[1715533164]

1715533164

[RichBC]

Yes it is actually hashing as reported by Antpool. That said I have not looked too closely just restarted it. The slight complication is that I run at 15GH & usually when it "goes slow" it is showing the standard speed of 5GH, so although I do not think this is the case, I had wondered if it had somehow picked up the default frequency from somewhere?

Will take a closer look next time it does this. Is there anything else I can look at to help with the problem?


Rich

you can ssh in and run cgminer-api to see what it says.. it would give you more of a direct stats then the ui or the pool..

if it shows the 5s at 15ghs but the avg at 5gsh then its just averaging wrong and everything is ok. you can restart it but its prob not needed.

i can show you a quick script that will restart if gsh goes low..

put this in a text and put it in /usr/bin and make it 0755 then run it with scheduled tasks running every 5 or 10 minutes.

###START
minhash=1400
SPEED=`cgminer-api | grep 'GHS av] => ' | cut -c 16- | tr -d '.'`

  if [ "$SPEED" -lt $minhash ]; then
    logger "cgminer was restarted for low hash"
    /etc/init.d/cgminer stop
    killall -9 cgminer
    sleep 1
    /etc/init.d/cgminer start
    exit 0;

fi
###END

what this will do is check the avg ghs, if its less then 14 it will restart.. you can change the min to whatever. no decimal.

Thanks but I need this spelling out in more detail if I am going to do it?

So I take the above & put it in a txt file?
Do I include ###START & END?
What do you mean by make it 0755? Is that the file name?
What text do I put in the scheduled tasks?
Can I leave the cron job there as well?

As you can see I have no idea what I am doing...

Rich

[aarons6]

Yes it is actually hashing as reported by Antpool. That said I have not looked too closely just restarted it. The slight complication is that I run at 15GH & usually when it "goes slow" it is showing the standard speed of 5GH, so although I do not think this is the case, I had wondered if it had somehow picked up the default frequency from somewhere?

Will take a closer look next time it does this. Is there anything else I can look at to help with the problem?


Rich

you can ssh in and run cgminer-api to see what it says.. it would give you more of a direct stats then the ui or the pool..

if it shows the 5s at 15ghs but the avg at 5gsh then its just averaging wrong and everything is ok. you can restart it but its prob not needed.

i can show you a quick script that will restart if gsh goes low..

put this in a text and put it in /usr/bin and make it 0755 then run it with scheduled tasks running every 5 or 10 minutes.

###START
minhash=1400
SPEED=`cgminer-api | grep 'GHS av] => ' | cut -c 16- | tr -d '.'`

  if [ "$SPEED" -lt $minhash ]; then
    logger "cgminer was restarted for low hash"
    /etc/init.d/cgminer stop
    killall -9 cgminer
    sleep 1
    /etc/init.d/cgminer start
    exit 0;

fi
###END

what this will do is check the avg ghs, if its less then 14 it will restart.. you can change the min to whatever. no decimal.

Thanks but I need this spelling out in more detail if I am going to do it?

So I take the above & put it in a txt file?
Do I include ###START & END?
What do you mean by make it 0755? Is that the file name?
What text do I put in the scheduled tasks?
Can I leave the cron job there as well?

As you can see I have no idea what I am doing...

Rich

its quite easy ill explain it..
you can take the lines from start to end, you can include the ### if you want.. not necessary tho and past it in a txt.. name it hashcheck.. by default windows will make it hashcheck.txt unless you turned off the showing file types option..


put that file in /usr/bin using wincsp.. right click that file rename it and take the .txt off and right click and hit properties and in the octal box type in 0755 and hit ok. this will make it executable.

now go in the R1 UI and where the scheduled tasks is under
*/3 *  *   *   *     /usr/bin/cgminer-monitor

put
*/5 *  *   *   *     /usr/bin/hashcheck

the /5 part is 5 minutes.. you can put /10 or /30 or whatever although i wouldn't make it less than 5.

you shouldnt have to reboot or anything just hit submit and it will start running.
let me know if it helps.

[RichBC]

its quite easy ill explain it..
you can take the lines from start to end, you can include the ### if you want.. not necessary tho and past it in a txt.. name it hashcheck.. by default windows will make it hashcheck.txt unless you turned off the showing file types option..


put that file in /usr/bin using wincsp.. right click that file rename it and take the .txt off and right click and hit properties and in the octal box type in 0755 and hit ok. this will make it executable.

now go in the R1 UI and where the scheduled tasks is under
*/3 *  *   *   *     /usr/bin/cgminer-monitor

put
*/5 *  *   *   *     /usr/bin/hashcheck

the /5 part is 5 minutes.. you can put /10 or /30 or whatever although i wouldn't make it less than 5.

you shouldnt have to reboot or anything just hit submit and it will start running.
let me know if it helps.

OK that is all done & running, will let you know how it goes. Thanks for spelling it out...

So just for general education.   I think I can see that this runs cgminer-api then picks up on the Average GH looks for it to be >= but then I get lost on that line?

Is there a guide somewhere to this "language"?

Rich

[aarons6]

OK that is all done & running, will let you know how it goes. Thanks for spelling it out...

So just for general education.   I think I can see that this runs cgminer-api then picks up on the Average GH looks for it to be >= but then I get lost on that line?

Is there a guide somewhere to this "language"?

Rich

its shell or bash or sh.. those are all the same.

you are right.. its running cgminer-api then its looking for the line GHS av if you went in SSH it would look like this

Code:

[STATUS] =>
(
   [STATUS] => S
   [When] => 1447936121
   [Code] => 11
   [Msg] => Summary
   [Description] => cgminer 4.8.0
)
[SUMMARY] =>
(
   [0] => SUMMARY
   [Elapsed] => 509
   [GHS 5s] => 5.45
   [GHS av] => 5.12
   [Found Blocks] => 0
   [Getworks] => 20
   [Accepted] => 34
   [Rejected] => 0
   [Hardware Errors] => 0
   [Utility] => 4.01
   [Discarded] => 286
   [Stale] => 0
   [Get Failures] => 0
   [Local Work] => 1374
   [Remote Failures] => 0
   [Network Blocks] => 2
   [Total MH] => 2606530.0000
   [Work Utility] => 71.55
   [Difficulty Accepted] => 544.00000000
   [Difficulty Rejected] => 0.00000000
   [Difficulty Stale] => 0.00000000
   [Best Share] => 801
   [Device Hardware%] => 0.0000
   [Device Rejected%] => 0.0000
   [Pool Rejected%] => 0.0000
   [Pool Stale%] => 0.0000
   [Last getwork] => 1447936121
)

then what it does is strip the words, and the decimal, so it gives a solid number.. 512 in my case..
then it checks to see if its less than your preset.. -lt

this is a good site for info
http://linuxcommand.org/
[/code]

[RichBC]

Brilliant , thank you. There are obviously a neat set of things that could be done with scripts running on miners.   You often get people wanting to check or restart or change things under different circumstances. I am surprised that there is not a set of handy scripts around.

Couple of things. The R1 has a tab for scheduled tasks, however I have not seen that tab on an S5 is there another way of setting one up? Also is it possible to do things like check for HW errors and then adjust the frequency up or down?


Rich

[aarons6]

Brilliant , thank you. There are obviously a neat set of things that cold be done with scripts running on miners.   You often get people wanting to check or restart or change things under different circumstances. I am surprised that there is not a set of handy scripts around.

Couple of things. The R1 has a tab for scheduled tasks, however I have not seen that tab on an S5 is there another way of setting one up? Also is it possible to do things like check for HW errors and then adjust the frequency up or down?


Rich

the s5 doesnt have crontab..

so you would have to make a script to go in init.d.. making one to adjust the speed is possible. you would have to make it smart, checking both hardware errors and temp..

[RichBC]

Brilliant , thank you. There are obviously a neat set of things that cold be done with scripts running on miners.   You often get people wanting to check or restart or change things under different circumstances. I am surprised that there is not a set of handy scripts around.

Couple of things. The R1 has a tab for scheduled tasks, however I have not seen that tab on an S5 is there another way of setting one up? Also is it possible to do things like check for HW errors and then adjust the frequency up or down?


Rich

the s5 doesnt have crontab..

so you would have to make a script to go in init.d.. making one to adjust the speed is possible. you would have to make it smart, checking both hardware errors and temp..

So yes I understand you would want to check HW error & temp but in principal it could be done including sending a command to change the frequency?

Rich

[aarons6]

Brilliant , thank you. There are obviously a neat set of things that cold be done with scripts running on miners.   You often get people wanting to check or restart or change things under different circumstances. I am surprised that there is not a set of handy scripts around.

Couple of things. The R1 has a tab for scheduled tasks, however I have not seen that tab on an S5 is there another way of setting one up? Also is it possible to do things like check for HW errors and then adjust the frequency up or down?


Rich

the s5 doesnt have crontab..

so you would have to make a script to go in init.d.. making one to adjust the speed is possible. you would have to make it smart, checking both hardware errors and temp..

So yes I understand you would want to check HW error & temp but in principal it could be done including sending a command to change the frequency?

Rich

yeah the freq on the s5 is adjusted by a cgminer.conf file.. it would be easy to make a few, like say 350(stock), 400, 425, and then have the script switch them out if the temp or hwe is too high but prefer the highest one.

you would have to make it count the hwe over a certain time.. so if you made the script run every 10 minutes you can put it in if hwe is 100 more then last time.. then lower it.. or whatever.

the script part is easy..

[RichBC]

After a couple of Hours I have had my first Stop, not go slow, and something is not right as there was not a restart. Here is the output from cgminer-api

Code:

   [0] => SUMMARY
   [Elapsed] => 11806
   [GHS 5s] => 0.00
   [GHS av] => 11.30
   [Found Blocks] => 0
   [Getworks] => 239
   [Accepted] => 1936
   [Rejected] => 1
   [Hardware Errors] => 6
   [Utility] => 9.84
   [Discarded] => 6315
   [Stale] => 0
   [Get Failures] => 0
   [Local Work] => 139750
   [Remote Failures] => 0
   [Network Blocks] => 17
   [Total MH] => 133422325.0000
   [Work Utility] => 157.91
   [Difficulty Accepted] => 30976.00000000
   [Difficulty Rejected] => 16.00000000
   [Difficulty Stale] => 0.00000000
   [Best Share] => 64181
   [Device Hardware%] => 0.0193
   [Device Rejected%] => 0.0515
   [Pool Rejected%] => 0.0516
   [Pool Stale%] => 0.0000
   [Last getwork] => 1447945808

This is from the System Log, you can see that the job is running.

Code:

Thu Nov 19 15:14:01 2015 cron.info crond[844]: crond: USER root pid 5683 cmd /usr/bin/hashcheck
Thu Nov 19 15:15:01 2015 cron.info crond[844]: crond: USER root pid 5690 cmd /usr/bin/hashcheck
Thu Nov 19 15:15:04 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:16:01 2015 cron.info crond[844]: crond: USER root pid 5697 cmd /usr/bin/hashcheck
Thu Nov 19 15:16:08 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:16:48 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:17:01 2015 cron.info crond[844]: crond: USER root pid 5705 cmd /usr/bin/hashcheck

Either I have done something wrong or the code is not right?


Rich

[Mikestang]

as soon as you plug it in its setup to go online with bitmains id. so im pretty sure that number is pretty valid as the total..

Unless you buy the R1 from someone like CrazyGuy who has already modified the firmware for you.  That's what I plan to do, my R1 will never be seen by Bitmain on their pool so it's possible there are many more out there than reported.

[RichBC]

as soon as you plug it in its setup to go online with bitmains id. so im pretty sure that number is pretty valid as the total..

Unless you buy the R1 from someone like CrazyGuy who has already modified the firmware for you.  That's what I plan to do, my R1 will never be seen by Bitmain on their pool so it's possible there are many more out there than reported.

You are correct CG first pool is Solo at ckpool with his Bitcoin Address  so it will not be logged. However these ought to be a tiny fraction of the Total R1?

Rich

[notlist3d]

as soon as you plug it in its setup to go online with bitmains id. so im pretty sure that number is pretty valid as the total..

Unless you buy the R1 from someone like CrazyGuy who has already modified the firmware for you.  That's what I plan to do, my R1 will never be seen by Bitmain on their pool so it's possible there are many more out there than reported.

CrazyGuy does have it where you could use antpool if you want to see some stat's to play with on antpool site on that antrouter tab.   But you are right if you stick with ck solo it would never touch on his sales with custom firmware.  I am very happy with my purchase from him.

I really am surprised bitmain did not do a batch 2 yet.  Guess we will see I'm happy if it turns into more collector item with only around 200 out there... but I would think they would want to sell more though.

[alh]

While I'll admit this is speculation on my part, I'll bet that the profit margins on the S7 are way higher than the margins on the R1 router. And by margin I mean both as a percentage, and in absolute dollars/BTC (obviously).

Does anybody know what the peak was in terms of R1's that logged into Antpool to do their duty for the collective (i.e. help mining for Antminer). My recollection is that an R1 user/owner has some insight into that information.

[notlist3d]

While I'll admit this is speculation on my part, I'll bet that the profit margins on the S7 are way higher than the margins on the R1 router. And by margin I mean both as a percentage, and in absolute dollars/BTC (obviously).

Does anybody know what the peak was in terms of R1's that logged into Antpool to do their duty for the collective (i.e. help mining for Antminer). My recollection is that an R1 user/owner has some insight into that information.

I think you hit the nail on the head they make more selling S7's. These sold for what 30 bucks plus shipping (shipping was expensive).   That is much cheaper then I expected though.  I thought they would put at least double this.  But they made it for a fair price. They made money but even if they doubled money i'm willing to bet it's not even close to S7 profits.

It's actually not even on the front page on items being sold - https://www.bitmaintech.com/product.htm .   So either they dropped it... or they are working on new version.  I think most likely is they dropped it for the moment and focus more of S7's like you mentioned.   I hope they go back sometime to it as it was a pretty neat device.

With crazy guy's custom firmware it is a amazing unit.  I love the U3 resetting, I don't have to unplug multiple times a day which is good.

[aarons6]

After a couple of Hours I have had my first Stop, not go slow, and something is not right as there was not a restart. Here is the output from cgminer-api

Code:

   [0] => SUMMARY
   [Elapsed] => 11806
   [GHS 5s] => 0.00
   [GHS av] => 11.30
   [Found Blocks] => 0
   [Getworks] => 239
   [Accepted] => 1936
   [Rejected] => 1
   [Hardware Errors] => 6
   [Utility] => 9.84
   [Discarded] => 6315
   [Stale] => 0
   [Get Failures] => 0
   [Local Work] => 139750
   [Remote Failures] => 0
   [Network Blocks] => 17
   [Total MH] => 133422325.0000
   [Work Utility] => 157.91
   [Difficulty Accepted] => 30976.00000000
   [Difficulty Rejected] => 16.00000000
   [Difficulty Stale] => 0.00000000
   [Best Share] => 64181
   [Device Hardware%] => 0.0193
   [Device Rejected%] => 0.0515
   [Pool Rejected%] => 0.0516
   [Pool Stale%] => 0.0000
   [Last getwork] => 1447945808

This is from the System Log, you can see that the job is running.

Code:

Thu Nov 19 15:14:01 2015 cron.info crond[844]: crond: USER root pid 5683 cmd /usr/bin/hashcheck
Thu Nov 19 15:15:01 2015 cron.info crond[844]: crond: USER root pid 5690 cmd /usr/bin/hashcheck
Thu Nov 19 15:15:04 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:16:01 2015 cron.info crond[844]: crond: USER root pid 5697 cmd /usr/bin/hashcheck
Thu Nov 19 15:16:08 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:16:48 2015 daemon.warn dnsmasq-dhcp[1133]: no address range available for DHCP request via wlan0
Thu Nov 19 15:17:01 2015 cron.info crond[844]: crond: USER root pid 5705 cmd /usr/bin/hashcheck

Either I have done something wrong or the code is not right?


Rich

the avg has to drop down, not the 5s.
you can change it to 5s tho if you need.


edit.. i wouldnt run it every minute.. id do it like 5 or 10.

[RichBC]

The average is 11.3 which is < 14 in the above.

Aside from that my whole setup stopped hashing and there was no restart. I restarted manually still no hash, reboot still no hash, power down still no hash, hard reset still not working.

Eventually realised that Antpool had gone down. Duh... so have had now to rebuild everything. Anyway it's all back and running and I think the script is working now but will have to wait & see.

Yes agreed it's running too frequently but lets me check quicker.


Rich

[citronick]

as soon as you plug it in its setup to go online with bitmains id. so im pretty sure that number is pretty valid as the total..

Unless you buy the R1 from someone like CrazyGuy who has already modified the firmware for you.  That's what I plan to do, my R1 will never be seen by Bitmain on their pool so it's possible there are many more out there than reported.

CrazyGuy does have it where you could use antpool if you want to see some stat's to play with on antpool site on that antrouter tab.   But you are right if you stick with ck solo it would never touch on his sales with custom firmware.  I am very happy with my purchase from him.

I really am surprised bitmain did not do a batch 2 yet.  Guess we will see I'm happy if it turns into more collector item with only around 200 out there... but I would think they would want to sell more though.

can someone post the link to crazyguy's firmware? install guide?

[aarons6]

look what i found here.. exactly what i thought.. r1 is running out of memory..

Code:

Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] cg@Watchdog invoked oom-killer: gfp_mask=0x2d2, order=0, oom_score_adj=0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] CPU: 0 PID: 1379 Comm: cg@Watchdog Not tainted 3.10.12 #36
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Stack : 00000006 00000000 00000000 00000000 00000000 00000000 803a2a8e 0000003a
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 	  838a6758 802d8370 80382a04 8032cf7b 00000563 000002d2 00000000 00000000
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 	  8032d050 80078f68 80330000 80076a7c 00000000 00000000 802d9c38 83075864
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 	  00075864 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 830757f0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 	  ...
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Call Trace:
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8006e27c>] show_stack+0x48/0x70
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800b8bbc>] dump_header.isra.16+0x4c/0x138
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800b8f08>] oom_kill_process+0xd4/0x3b0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800b9648>] out_of_memory+0x290/0x2ec
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800bc6c4>] __alloc_pages_nodemask+0x558/0x63c
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800dad2c>] __vmalloc_node_range+0x120/0x214
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800dae50>] __vmalloc_node+0x30/0x3c
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800dae88>] vmalloc+0x2c/0x38
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<80187c90>] LzmaDecode+0x124/0x4ec
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8014a870>] jffs2_lzma_decompress+0x60/0x94
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<80138230>] jffs2_decompress+0xbc/0x134
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8013b644>] jffs2_read_dnode+0x2a4/0x384
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8013b848>] jffs2_read_inode_range+0x124/0x1b0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<80139c28>] jffs2_do_readpage_nolock+0x64/0xe8
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8013a010>] jffs2_do_readpage_unlock+0x18/0x3c
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<8013a06c>] jffs2_readpage+0x38/0x5c
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800bf42c>] __do_page_cache_readahead+0x1b8/0x234
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800bf7b0>] ra_submit+0x28/0x34
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800b80dc>] filemap_fault+0x198/0x3e8
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800ce1d0>] __do_fault+0xcc/0x444
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800d1128>] handle_pte_fault+0x32c/0x6dc
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<800d1588>] handle_mm_fault+0xb0/0xdc
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<80071904>] do_page_fault+0x110/0x354
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] [<80060820>] ret_from_exception+0x0/0xc
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Mem-Info:
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Normal per-cpu:
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] CPU    0: hi:   18, btch:   3 usd:  15
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] active_anon:12086 inactive_anon:33 isolated_anon:0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000]  active_file:11 inactive_file:13 isolated_file:0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000]  unevictable:0 dirty:0 writeback:0 unstable:0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000]  free:241 slab_reclaimable:297 slab_unreclaimable:1087
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000]  mapped:2 shmem:160 pagetables:106 bounce:0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000]  free_cma:0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Normal free:964kB min:988kB low:1232kB high:1480kB active_anon:48344kB inactive_anon:132kB active_file:44kB inactive_file:52kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:65536kB managed:61148kB mlocked:0kB dirty:0kB writeback:0kB mapped:8kB shmem:640kB slab_reclaimable:1188kB slab_unreclaimable:4348kB kernel_stack:560kB pagetables:424kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:197 all_unreclaimable? yes
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] lowmem_reserve[]: 0 0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Normal: 19*4kB (M) 97*8kB (M) 7*16kB (UM) 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 964kB
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 184 total pagecache pages
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 0 pages in swap cache
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Swap cache stats: add 0, delete 0, find 0/0
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Free swap  = 0kB
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] Total swap = 0kB
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 16384 pages RAM
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 1037 pages reserved
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 132612 pages shared
Thu Nov 19 22:48:13 2015 kern.warn kernel: [14263.730000] 14793 pages non-shared
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ pid ]   uid  tgid total_vm      rss nr_ptes swapents oom_score_adj name
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  651]     0   651      227       22       5        0             0 ubusd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  797]     0   797      371       47       4        0             0 netifd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  841]     0   841      379       22       3        0             0 crond
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  852]     0   852      289       18       4        0             0 dropbear
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  881]     0   881      375       18       4        0             0 udhcpc
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [  894]     0   894      291       26       3        0             0 uhttpd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1113]     0  1113      399       26       3        0             0 wpa_supplicant
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1132] 65534  1132      239       21       4        0             0 dnsmasq
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1215]     0  1215      375       18       5        0             0 ntpd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1343]     0  1343    20422    10865      22        0             0 cgminer-u3
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1435]     0  1435      242       34       5        0         -1000 udevd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1461]     0  1461     7690      666      11        0             0 cgminer
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 1465]     0  1465      242       38       5        0         -1000 udevd
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3445]     0  3445      374       17       3        0             0 sh
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3446]     0  3446      373       16       4        0             0 exe
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3447]     0  3447      374       17       4        0             0 exe
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3448]     0  3448      193        9       3        0             0 cgminer-api
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3449]     0  3449      372       11       3        0             0 grep
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3450]     0  3450      372       14       3        0             0 cut
Thu Nov 19 22:48:13 2015 kern.info kernel: [14263.730000] [ 3451]     0  3451      372       10       3        0             0 tr
Thu Nov 19 22:48:13 2015 kern.err kernel: [14263.730000] Out of memory: Kill process 1343 (cgminer-u3) score 680 or sacrifice child
Thu Nov 19 22:48:13 2015 kern.err kernel: [14263.730000] Killed process 1343 (cgminer-u3) total-vm:81688kB, anon-rss:43460kB, file-rss:0kB

[RichBC]

I have made a couple of tweeks to hashcheck script. I found that when the R1 has been running for a while the Average GH/s decays very slowly when the miner stops which significantly delays the restart, so I am monitoring instead the 5s and have set the level to 10000 (Am hashing at 16.5GH/s) Have also put in some logger lines which output the speed, interval set at 2 mins.

So it's been running now for a couple of Hours without a stop, we will see what happens?

So on the running out of memory where are you seeing that info and what happens? Any thoughts on a fix?


Rich