I've a BBW running 24/7 since December with: BeagleBoard.org Debian Image 2015-11-12, kernel 4.1.18-ti-r49 #1 SMP PREEMPT Fri Feb 26 00:12:54 UTC 2016 armv7l GNU/Linux
It is on a UPS and I've had exactly one lock up, happened last week -- actually looked like it had powered down, all LEDs off. We also had a thunderstorm roll through about when its last heartbeat message came, but the UPS never "beeped" nor did the lights flicker (unusual, this generally means it was only a very minor thunderstorm as things go around here). I posted a thread on it when it happened. What was weird was the reset button or unplugging and re-plugging the 5V "barrel connector" power-supply didn't restart it, so I thought it'd fried. I left it unplugged while checking some other parts of the system and when I when back to it 10+ minutes later it booted right up when I plugged it in -- almost like some kind of thermal overload, but nothing is even warm to the touch on the board. It still running 24/7 since, another small thunderstorm is rolling through as I type this, I'll see what if anything happens. On Thursday, May 19, 2016 at 12:08:17 PM UTC-5, Super Twang wrote: > > > Robert, > > You make a good point. Maybe a brownout/blackout caused it. We had a > tstorm roll through here last night, and I’m not running power from a UPS, > I wish there was someway to catch a spontaneous power loss in the logs so > I could differentiate that out, but the syslog shows nothing in these > cases. I’ll try again with (more) stable power. > > Is it normal for the buffer count to steadily increase like I’m seeing? > Could it be my own stat logging that is causing it? Maybe I’m doing > something and just not seeing it straight, but I’m trying to have as > minimal an impact as I can on the running system. > > Thanks for your thoughts. > > Best, > ST > > PS. Out of curiosity, how many machines are in your testing farm? Do you > have a photo? I’ve got a whopping 2. :) > > > On May 19, 2016, at 11:37 AM, Robert Nelson <robert...@gmail.com > <javascript:>> wrote: > > That's odd... > > http://rcn-ee.homeip.net:81/farm/uptime/test-bbb-6.log > > Double check your power supply.. > > On May 19, 2016 11:34 AM, "Super Twang" <super...@gmail.com <javascript:>> > wrote: > > > > > > PS. Is this the best place to put bug/issue reports? I believe this > relates to the core image & kernel of a Jessie LTS release. > > > > --- > > > > I'm seeing semi-regular spontaneous reboots on the BBB. Here's what I'm > seeing: > > > > Running 'top', the "Mem Buffers" field, (4th line, far right), is > steadily increasing until the system reboots at around (41180 buffers). > > > > Anyhow, since I'm going after max stability, this is of concern to me, > so I've been doing a little testing. > > > > I'm not entirely sure what I'm looking for, but since I'm not running > any of my own apps in this setup, I'm hoping the results might be useful > for someone else? > > > > > > I have logs available from > > > > vmstat -s K -n 1 >> logfile > > > > as well as a screengrab at time of fail of > > > > 'top' > > > > and > > > > 'slabtop' > > > > Let me know if you have any insights or would like the logs. I still > have the image and am happy to run other tests if it'll help. > > In the meantime, I'm going to revert to a Wheezy console image and hope > to find more stability there. > > > > Best, > > Dave > > > > --------- setup is below > > > > Setup: > > > > Running from image: RCN's jessie 8.4 console image of 4/7/2016 (with > some packages removed, and a developer environment installed) > > > > Kernel: 4.4.9 -bone-rt-r10 kernel > > > > Running from SDCard > > > > ssh'ed in via DCHP connected ethernet, most recently capturing vmstat to > a log file. > > > > > > > > PS. Under a different testing scenario where I am running my own http > server/app, I see strange behavior too. I can say that my app has been > extensively tested for leaks, etc, BUT can't rule out that it is my own app > causing the issues when I am running it. Nonetheless, maybe the > strangeness will help dignose the above issue, so I'm including it here. > > > > The strangeness I'm seeing is this: There is a 24 hour recurring > pattern to the reboots. In the log below, I'm the number to the right of > "Startup" is "seconds since last startup". > > > > Any idea what this might be??? Logfile rotation? Some kind of cron-job > interference? > > > > 2016-05-17 07:36:14: Startup 15667 31504 * > > > > 2016-05-17 11:57:11: Startup 15657 * * > > > > 2016-05-17 14:21:10: Startup 8639 * * > > > > 2016-05-17 16:20:27: Startup 7157 * * > > > > 2016-05-17 17:22:41: Startup 3734 * * > > > > 2016-05-17 18:23:23: Startup 3642 * * > > > > 2016-05-17 19:24:06: Startup 3643 * * > > > > 2016-05-17 20:25:33: Startup 3687 * * > > > > 2016-05-17 21:26:16: Startup 3643 * * > > > > 2016-05-17 22:27:00: Startup 3644 * * > > > > 2016-05-17 23:27:41: Startup 3641 * * > > > > 2016-05-18 00:28:24: Startup 3643 * * > > > > 2016-05-18 01:30:36: Startup 3732 * * > > > > 2016-05-18 02:29:06: Startup 3510 * * > > > > 2016-05-18 03:17:03: Startup 2877 * * > > > > 2016-05-18 03:21:32: Startup 269 * * > > > > 2016-05-18 03:25:59: Startup 267 * * > > > > 2016-05-18 03:30:27: Startup 268 * * > > > > 2016-05-18 03:34:54: Startup 267 * * > > > > 2016-05-18 07:25:51: Startup 13857 * * > > > > 2016-05-18 11:22:51: Startup 14220 * * > > > > 2016-05-18 12:51:18: Startup 5307 * * > > > > 2016-05-18 13:51:59: Startup 3641 * * > > > > 2016-05-18 14:54:13: Startup 3734 * * > > > > 2016-05-18 15:55:42: Startup 3689 * * > > > > 2016-05-18 16:56:25: Startup 3643 * * > > > > 2016-05-18 17:57:54: Startup 3689 * * > > > > > > -- > > For more options, visit http://beagleboard.org/discuss > > --- > > You received this message because you are subscribed to the Google > Groups "BeagleBoard" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to beagleboard...@googlegroups.com <javascript:>. > > To view this discussion on the web visit > https://groups.google.com/d/msgid/beagleboard/d9d73516-35b8-4673-96b5-e8dc9bef93d7%40googlegroups.com > . > > For more options, visit https://groups.google.com/d/optout. > > -- > For more options, visit http://beagleboard.org/discuss > --- > You received this message because you are subscribed to a topic in the > Google Groups "BeagleBoard" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/beagleboard/BTnO7AiHy8g/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > beagleboard...@googlegroups.com <javascript:>. > To view this discussion on the web visit > https://groups.google.com/d/msgid/beagleboard/CAOCHtYhLXvvT9ATJCir4et-N42duXzCPginVdWdZZ0nnp1sGqw%40mail.gmail.com > > <https://groups.google.com/d/msgid/beagleboard/CAOCHtYhLXvvT9ATJCir4et-N42duXzCPginVdWdZZ0nnp1sGqw%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > > > -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/237351b0-3f07-4435-a5c7-1198e5f53ac1%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
