vj Posted November 29, 2011 Posted November 29, 2011 I am getting these error messages in syslog. I recently had 1 GB memory, but, one ram went bad and am currently running on 512MB. I have ordered more 2GB RAM, but will take a few days to get it. I just wanted someone to take a look and tell me if the messages are only due to lack of memory and nothing else. Unraid is not crashing, it is working fine, but a little slower than before. A memtest on the existing 512 MB gave no errors. The only add-ons I have are unmenu, unraid-status-email (and related packages that it depends upon) and bwm-ng. Error message is as follows (shows up red in unmenu): ======================================== Nov 28 18:24:26 Tower kernel: SLUB: Unable to allocate memory on node -1 (gfp=0x20) Nov 28 18:24:26 Tower kernel: cache: kmalloc-4096, object size: 4096, buffer size: 4096, default order: 3, min order: 0 Nov 28 18:24:26 Tower kernel: node 0: slabs: 519, objs: 1030, free: 0 Nov 28 18:24:26 Tower kernel: swapper: page allocation failure. order:0, mode:0x4020 Nov 28 18:24:26 Tower kernel: Pid: 0, comm: swapper Not tainted 2.6.32.9-unRAID #8 (Errors) Nov 28 18:24:26 Tower kernel: Call Trace: (Errors) Nov 28 18:24:26 Tower kernel: [<c104d06f>] __alloc_pages_nodemask+0x3fb/0x42f (Errors) Nov 28 18:24:26 Tower kernel: [<c106841c>] __slab_alloc+0x13e/0x425 (Errors) Nov 28 18:24:26 Tower kernel: [<c1243051>] ? ip_local_deliver+0xb2/0x154 (Errors) Nov 28 18:24:26 Tower kernel: [<c1068d8f>] __kmalloc_track_caller+0x86/0xcf (Errors) Nov 28 18:24:26 Tower kernel: [<c1229a69>] ? dev_alloc_skb+0x14/0x29 (Errors) Nov 28 18:24:26 Tower kernel: [<c1229a69>] ? dev_alloc_skb+0x14/0x29 (Errors) Nov 28 18:24:26 Tower kernel: [<c1229709>] __alloc_skb+0x50/0x119 (Errors) Nov 28 18:24:26 Tower kernel: [<c1229a69>] dev_alloc_skb+0x14/0x29 (Errors) Nov 28 18:24:26 Tower kernel: [<dc7295e4>] nv_alloc_rx_optimized+0x3e/0x198 [forcedeth] (Errors) Nov 28 18:24:26 Tower kernel: [<dc72df3d>] nv_napi_poll+0x48b/0x49e [forcedeth] (Errors) Nov 28 18:24:26 Tower kernel: [<c1002f29>] ? common_interrupt+0x29/0x30 (Errors) Nov 28 18:24:26 Tower kernel: [<c12312aa>] net_rx_action+0x57/0x102 (Errors) Nov 28 18:24:26 Tower kernel: [<c1028261>] __do_softirq+0x84/0xf8 (Errors) Nov 28 18:24:26 Tower kernel: [<c10282fb>] do_softirq+0x26/0x2b (Errors) Nov 28 18:24:26 Tower kernel: [<c1028556>] irq_exit+0x29/0x2b (Errors) Nov 28 18:24:26 Tower kernel: [<c10042c5>] do_IRQ+0x80/0x96 (Errors) Nov 28 18:24:26 Tower kernel: [<c1002f29>] common_interrupt+0x29/0x30 (Errors) Nov 28 18:24:26 Tower kernel: [<c1008160>] ? default_idle+0x2d/0x42 (Errors) Nov 28 18:24:26 Tower kernel: [<c100837c>] c1e_idle+0xc9/0xce (Errors) Nov 28 18:24:26 Tower kernel: [<c1001a14>] cpu_idle+0x3a/0x4e (Errors) Nov 28 18:24:26 Tower kernel: [<c128a8bf>] rest_init+0x53/0x55 (Errors) Nov 28 18:24:26 Tower kernel: [<c13f580c>] start_kernel+0x27b/0x280 (Errors) Nov 28 18:24:26 Tower kernel: [<c13f5091>] i386_start_kernel+0x91/0x96 (Errors) ================================== Top 5 lines of "top" command: ================================== top - 18:40:28 up 19:21, 1 user, load average: 0.16, 0.08, 0.08 Tasks: 67 total, 2 running, 65 sleeping, 0 stopped, 0 zombie Cpu(s): 0.7%us, 0.7%sy, 0.0%ni, 94.0%id, 4.3%wa, 0.0%hi, 0.3%si, 0.0%st Mem: 448564k total, 442812k used, 5752k free, 39404k buffers Swap: 0k total, 0k used, 0k free, 345300k cached =================================== Output of free command: ======================== root@Tower:~# free total used free shared buffers cached Mem: 448564 442652 5912 0 41032 342740 -/+ buffers/cache: 58880 389684 Swap: 0 0 0 =========================== Thanks! VJ syslog-2011-11-28.txt
prostuff1 Posted November 29, 2011 Posted November 29, 2011 How long did you run memtest? at least over night is needed and preferably for 24 hours
vj Posted November 29, 2011 Author Posted November 29, 2011 I just ran it for 15 minutes until I got the message "****Pass complete, no errors, press ESC to exit ******. I will run it overnight today and update you. VJ
vj Posted November 29, 2011 Author Posted November 29, 2011 Ran it for 7.5 hours and got 1 error on screen, though error count says 4. I have attached a screenshot of the test. I went ahead and added a cache drive and a swapfile to my array. I also saw some new errors in my syslog. Seems to be related to the cache drive that I added last night. Did not have those yesterday in the syslog. I have attached the latest syslog too. Thanks! VJ syslog-2011-11-29.txt
prostuff1 Posted November 29, 2011 Posted November 29, 2011 do not change/do anything until you fix that memory error. Best way to do that is verify voltage and ram settings in BIOS and if those are correct then you will need to replace the ram sticks altogether. There should never be any errors returned from the memtest. If you don't fix it now you risk data corruption now and down the road.
vj Posted November 29, 2011 Author Posted November 29, 2011 Got it. Will check BIOS settings tonite. Thanks! VJ
vj Posted November 30, 2011 Author Posted November 30, 2011 my current memory is CORSAIR XMS2 512MB 240-Pin DDR2 SDRAM DDR2 675 (PC2 5400) Desktop Memory Model CM2X512-5400C4 (http://www.newegg.com/Product/Product.aspx?Item=N82E16820145538). I am not sure what settings I need to look/change in my BIOS. I have attached some screenshots of my BIOS. I see the following as different: The above website shows timing as 4-4-4-12, whereas in memtest it shows up as 5-5-5-15. Any ideas? Let me know what more information you will need. Thanks, VJ
vj Posted December 2, 2011 Author Posted December 2, 2011 I decided to reset my CMOS settings in the BIOS and since then the errors in the syslog have gone away. Maybe I set something which I was not supposed to while going through each option. But, the memtest still shows one error after running it for 8 hours. So, I will replace them when I get the new memory and check and see if those gives any errors. VJ
vj Posted December 3, 2011 Author Posted December 3, 2011 Got my new memory. 2 x 1 GB Kingston DDR2 800 (KVR800D2N6). These memory seemed to fit in easily in the memory slots compared to the old ones that I used and that made me feel much better putting them in. I started running memtest and it has been running for nearly 14 hours without any errors. Will run it for the another 24 hours. VJ
prostuff1 Posted December 3, 2011 Posted December 3, 2011 Got my new memory. 2 x 1 GB Kingston DDR2 800 (KVR800D2N6). These memory seemed to fit in easily in the memory slots compared to the old ones that I used and that made me feel much better putting them in. I started running memtest and it has been running for nearly 14 hours without any errors. Will run it for the another 24 hours. VJ Good to hear!! after that is complete you should be able to go back to using your unRAID machine normally.
vj Posted December 5, 2011 Author Posted December 5, 2011 No memory errors after running memtest for nearly 40 hours. Everything seems to be working fine for the last 2 days. No errors in syslog at all. Thanks for all your help. VJ
Recommended Posts
Archived
This topic is now archived and is closed to further replies.