Vm_page_unwire panic on Alix 2D13 2.2 system
I had one of these a day or 2 ago, connected to the serial console and there was the "db>" prompt.
Now I just updated to the latest 2.2
2.2-BETA (i386) built on Fri Oct 24 12:19:41 CDT 2014 FreeBSD 10.1-RC3
and it has done this:
Installing Service Watchdog and its dependencies.Starting syslog...done. Starting CRON... done. Starting package Cron...done. Starting package Service Watchdog...done. pfSense (nanobsd) 2.2-BETA i386 Fri Oct 24 12:19:41 CDT 2014 Bootup complete FreeBSD/i386 (testoffice-rt-01.np.net.inf.org) (ttyu0) *** Welcome to pfSense 2.2-BETA-nanobsd (i386) on testoffice-rt-01 *** WANADSL (wan) -> vr0 -> v4/DHCP4: 192.168.1.116/24 LAN (lan) -> vr1 -> v4: 10.49.208.250/22 OPT1SUBISU (opt1) -> vr2 -> OPT2WIFI (opt2) -> ath0_wlan0 -> v4: 10.49.212.250/22 OPT3VLAN20 (opt3) -> vr1_vlan20 -> v4: 10.11.20.250/24 OPT4VLAN30 (opt4) -> vr1_vlan30 -> v4: 10.11.30.250/24 OPT5VLAN40 (opt5) -> vr1_vlan40 -> v4: 10.11.40.250/24 OPT6VLAN50 (opt6) -> vr1_vlan50 -> v4: 10.11.50.250/24 OPT7VLAN60 (opt7) -> vr1_vlan60 -> v4: 10.11.60.250/24 OPT8VLAN70 (opt8) -> vr1_vlan70 -> v4: 10.11.70.250/24 0) Logout (SSH only) 9) pfTop 1) Assign Interfaces 10) Filter Logs 2) Set interface(s) IP address 11) Restart webConfigurator 3) Reset webConfigurator password 12) pfSense Developer Shell 4) Reset to factory defaults 13) Upgrade from console 5) Reboot system 14) Disable Secure Shell (sshd) 6) Halt system 15) Restore recent configuration 7) Ping host 16) Restart PHP-FPM 8) Shell Enter an option: panic: vm_page_unwire: page 0xc2e3e350's wire count is zero cpuid = 0 KDB: enter: panic [ thread pid 8920 tid 100116 ] Stopped at kdb_enter+0x3d: movl $0,kdb_why db>
It seems to have started about the time of FreeBSD 10.1-RC3, but also might be just a coincidence.
Anyone else seeing similar "panic" stops?
As can be seen in the console menu, this system also has a WiFi card in it ath0_wlan0 which is the NetGate-supplied KIT-ALIX-5004MP-DUAL. Might or might not be a relevant factor.
System has run all night without panic.
I'm running a similar ALIX on 2.2. ath card in hostAP (plus some VAPs that come and go depending on what I'm testing at the time). No VLANs. Haven't hit any panics on it.
Did you get prompted to submit a crash report post-reboot? If so and you did, if you could let me know (PM is fine) what IP that came from, I can get someone to check it out.
Been away from home for a week, wondered why the OpenVPN from this system at home was down (my APU system was up). Connected a serial cable and there is the "db>" prompt.
Is there useful things I can type and output to post before doing "reboot"?
It should generate a crash dump, which it'll prompt to submit on next login to the web interface. That'd have everything you could get at the db> prompt. Did it not seem to generate a crash dump?
Running 'bt' from the db> prompt would get the most useful info.
It is an Alix nanoBSD, so I guess the dump is in /var somewhere and does not survive reboot. I had already (accidentally) rebooted it. If/when it panics again I will get more output. Since no-one else is reporting anything, maybe this Alix has some dodgy hardware or…
Since ALIX lacks swap space (unless you've done a custom full install somehow), it will dump the crash report to the serial console and then auto-reboot in most cases. If it doesn't it would sit at a db> prompt. Make sure you have a large scrollback buffer or are recording the console session if you leave it open and waiting.
A week or two ago I'd lose contact with my test ALIX now and then but didn't witnessed a crash. I've been watching the serial console since doing my last update hoping to catch a crash but haven't seen anything and it's been working perfectly. Go figure. :-)
This Alix at home is happily up and running at the moment, I just logged into the webGUI over its site-to-site OpenVPN from work:
2.2-BETA (i386) built on Mon Nov 03 12:43:20 CST 2014 FreeBSD 10.1-RC4 Uptime 2 Days 00 Hour 48 Minutes 02 Seconds
I also have no idea how to induce a crash. I haven't seen one since 10.1-RC4 came along.
As we say in Nepal, के गर्ने? What to do?
Wait and see is all you can do for now. It may have been an issue that was fixed (by us or by FreeBSD!)
Just to update - this has been running fine the last few weeks. I have upgraded every couple of days to test stuff. No panics, no system problems at all. I will assume something got fixed in a late FreeBSD 10.1-RC.