Unable to allocate cache

[pirx73]

During router boot this keeps repeating:

Aug 31 21:45:06 kernel: SLUB: Unable to allocate memory on node -1 (gfp=0x21)
Aug 31 21:45:06 kernel: cache: kmalloc_dma-4096, object size: 4096, buffer size: 4096, default order: 3, min order: 0
Aug 31 21:45:06 kernel: node 0: slabs: 3829, objs: 30625, free: 0
Aug 31 21:45:06 kernel: swapper: page allocation failure. order:0, mode:0x4021
Aug 31 21:45:06 kernel: [<c0043ff8>] (unwind_backtrace+0x0/0xf8) from [<c009e84c>] (__alloc_pages_nodemask+0x61c/0x698)
Aug 31 21:45:06 kernel: [<c009e84c>] (__alloc_pages_nodemask+0x61c/0x698) from [<c0377764>] (__slab_alloc+0x794/0x814)
Aug 31 21:45:06 kernel: [<c0377764>] (__slab_alloc+0x794/0x814) from [<c0377998>] (__kmalloc_track_caller+0xf8/0x108)
Aug 31 21:45:06 kernel: [<c0377998>] (__kmalloc_track_caller+0xf8/0x108) from [<c0379f88>] (__alloc_skb+0x54/0x104)
Aug 31 21:45:06 kernel: [<c0379f88>] (__alloc_skb+0x54/0x104) from [<c01d8ea4>] (osl_alloc_skb+0x18/0x30)
Aug 31 21:45:06 kernel: [<c01d8ea4>] (osl_alloc_skb+0x18/0x30) from [<c03891f0>] (osl_pktget+0x4c/0x154)
Aug 31 21:45:06 kernel: [<c03891f0>] (osl_pktget+0x4c/0x154) from [<c038876c>] (_dma_rxfill+0x174/0x3d4)
Aug 31 21:45:06 kernel: [<c038876c>] (_dma_rxfill+0x174/0x3d4) from [<bf015bb4>] (chiprxfill+0x14/0x18 [et])
Aug 31 21:45:06 kernel: unwind: Index not found bf015bb4
Aug 31 21:45:06 kernel: Mem-info:
Aug 31 21:45:06 kernel: DMA per-cpu:
Aug 31 21:45:06 kernel: CPU 0: hi: 42, btch: 7 usd: 39
Aug 31 21:45:06 kernel: CPU 1: hi: 42, btch: 7 usd: 41
Aug 31 21:45:06 kernel: Normal per-cpu:
Aug 31 21:45:06 kernel: CPU 0: hi: 42, btch: 7 usd: 39
Aug 31 21:45:06 kernel: CPU 1: hi: 42, btch: 7 usd: 41
Aug 31 21:45:06 kernel: active_anon:2275 inactive_anon:2463 isolated_anon:0
Aug 31 21:45:06 kernel: active_file:1100 inactive_file:1324 isolated_file:0
Aug 31 21:45:06 kernel: unevictable:0 dirty:0 writeback:0 unstable:0
Aug 31 21:45:06 kernel: free:15498 slab_reclaimable:180 slab_unreclaimable:34122
Aug 31 21:45:06 kernel: mapped:1693 shmem:330 pagetables:192 bounce:0
Aug 31 21:45:06 kernel: DMA free:4124kB min:11020kB low:13772kB high:16528kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:130048kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:0kB slab_unreclaimable:122612kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes
Aug 31 21:45:06 kernel: lowmem_reserve[]: 0 109 109
Aug 31 21:45:06 kernel: Normal free:57868kB min:9456kB low:11820kB high:14184kB active_anon:9100kB inactive_anon:9852kB active_file:4400kB inactive_file:5296kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:111616kB mlocked:0kB dirty:0kB writeback:0kB mapped:6772kB shmem:1320kB slab_reclaimable:720kB slab_unreclaimable:13876kB kernel_stack:704kB pagetables:768kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
Aug 31 21:45:06 kernel: lowmem_reserve[]: 0 0 0
Aug 31 21:45:06 kernel: DMA: 3*4kB 2*8kB 0*16kB 4*32kB 0*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 4124kB
Aug 31 21:45:06 kernel: Normal: 51*4kB 60*8kB 152*16kB 47*32kB 58*64kB 29*128kB 11*256kB 4*512kB 4*1024kB 4*2048kB 7*4096kB = 57868kB
Aug 31 21:45:06 kernel: 2754 total pagecache pages
Aug 31 21:45:06 kernel: 0 pages in swap cache
Aug 31 21:45:06 kernel: Swap cache stats: add 0, delete 0, find 0/0
Aug 31 21:45:06 kernel: Free swap = 0kB
Aug 31 21:45:06 kernel: Total swap = 0kB
Aug 31 21:45:06 kernel: 65536 pages of RAM
Aug 31 21:45:06 kernel: 15875 free pages
Aug 31 21:45:06 kernel: 1609 reserved pages
Aug 31 21:45:06 kernel: 6737 slab pages
Aug 31 21:45:06 kernel: 7300 pages shared
Aug 31 21:45:06 kernel: 0 pages swap cached

free -m
total used free shared buffers cached
Mem: 255708 93436 162272 0 4108 31680
-/+ buffers/cache: 57648 198060
Swap: 0 0 0

 

[RMerlin]

Your router is running out of RAM.

 

[pirx73]

Thanks for the tip!
What could be causing this?
Traffic analyzer?
Because i don't have anything else running.
QoS is off and only jffs scripts i have are ledson.sh and ledsoff.sh

Sent from my FRD-L09 using Tapatalk

 

[RMerlin]

Thanks for the tip!
What could be causing this?
Traffic analyzer?
Because i don't have anything else running.
QoS is off and only jffs scripts i have are ledson.sh and ledsoff.sh

Sent from my FRD-L09 using Tapatalk

Do you have Download Master installed?

Otherwise, I don't see what could be using up your router's RAM so early at boot time.

 

[pirx73]

No DM is not installed, i don't even have USB flash attached.

 

[RMerlin]

See if anything is abnormally large in the ram-stored /tmp directory.

du -hx /tmp

 

[pirx73]

Not much:
1.3M /tmp

total

 

[RMerlin]

Not much:
1.3M /tmp

total

Looks fairly normal to me. Then I have no other idea what might be eating up your RAM, sorry.

 

[john9527]

I'm not sure it's not a hardware/addressing failure. If you expand his 'free' report everything looks quite normal with no memory shortage.

     total  used  free   shared buffers cached
Mem: 255708 93436 162272 0      4108    3168

 

[pirx73]

I'm not sure it's not a hardware/addressing failure.

Great...
I bought this router in may. Did not do any weird stuff with it. Just unpacked, updated to latest ASUS firmware. Performed reset to factory defaults.
Upgraded to Merlin's (i think 380.65 or .66 whatever was available in may), again reset to factory defaults and used it. No USB flash, no QoS enabled, only Traffic analyzer was running. I turned it off yesterday and deleted it's DB too.
This issue came up because i had to reboot router - it suddenly stopped working with wifi clients - they just dropped off and could not reconnect.
I attached syslog after rebooting it this morning

 

[pirx73]

Question is:
Does this fault clarify as RMA or i should not bother and it's fine to ignore it?

 

[mstombs]

The router doesn't crash during boot, your router stores some old syslog on jffs which gets added to on boot. The reboots start with time rolled back to Aug 1st, time not updated till ntp update runs

Aug 31 21:46:37 kernel: SLUB: Unable to allocate memory on node -1 (gfp=0x21)
Aug 31 21:46:37 kernel:   cache: kmalloc_dma-4096, object size: 4096, buffer size: 4096, default order: 3, min order: 0
Aug 31 21:46:37 kernel:   node 0: slabs: 3829, objs: 30625, free: 0
Aug  1 03:00:17 syslogd started: BusyBox v1.25.1
...
Aug  1 03:01:09 ntp: start NTP update
Aug 31 21:50:22 rc_service: ntp 867:notify_rc restart_upnp

The next ones look like a requested reboot?

Sep  1 11:44:58 iTunes: daemon is stopped
Sep  1 11:44:58 FTP_Server: daemon is stopped
Sep  1 11:44:59 Samba_Server: smb daemon is stopped
Sep  1 11:44:59 kernel: gro disabled
Sep  1 11:44:59 Timemachine: daemon is stopped
Sep  1 11:44:59 WEBDAV_Server: daemon is stopped
Sep  1 11:44:59 NAT_Tunnel: AAE Service is stopped
Sep  1 11:44:59 NAT_Tunnel: AAE Service is stopped
Sep  1 11:44:59 kernel: usbcore: deregistering interface driver cdc_mbim
Sep  1 11:44:59 kernel: usbcore: deregistering interface driver cdc_ncm
Sep  1 11:44:59 kernel: usbcore: deregistering interface driver qmi_wwan
Sep  1 11:44:59 kernel: usbcore: deregistering interface driver cdc_wdm
Sep  1 11:45:00 kernel: usbcore: deregistering interface driver rndis_host
Sep  1 11:45:00 kernel: usbcore: deregistering interface driver cdc_ether
Sep  1 11:45:00 kernel: usbcore: deregistering interface driver asix
Sep  1 11:45:00 kernel: usbcore: deregistering interface driver usblp
Aug  1 03:00:13 syslogd started: BusyBox v1.25.1
...
Aug  1 03:01:01 ntp: start NTP update
Sep  1 11:47:25 rc_service: ntp 871:notify_rc restart_upnp
...
Sep  2 12:19:08 kernel: usbcore: deregistering interface driver usblp
Aug  1 03:00:15 syslogd started: BusyBox v1.25.1
...
Aug  1 03:01:26 ntp: start NTP update
Sep  2 12:21:59 rc_service: ntp 830:notify_rc restart_upnp

What should the Linux kernel do when running low on ram?, often the OOM (out of memory) causes low priority big ram consuming tasks to be killed, and dnsmasq is a candidate since it runs as low privileged user after startup, but lack of dhcp and dns not good for a router!

 

[pirx73]

I dot not understand your point. I was asking what should i do to remedy this?
I do not have ANY extra services (itunes, aicloud, smb etc) running on the router, Traffic Analyzer is off and it's DB deleted.

 

[mstombs]

The log doesn't say what caused the problem on Aug 31 21:46:37 - my point was that the error messages are old, not re-occurring after boot.

 

[ColinTaylor]

The log doesn't say what caused the problem on Aug 31 21:46:37 - my point was that the error messages are old, not re-occurring after boot.

Or to put it another way; The log you provided showed the looping memory error on Aug 31. After that the router was deliberately rebooted at Sep 1 11:44:56 and Sep 2 12:19:05 but there are no errors in the log.

So the question is; was the error on the 31st a one-off or are you still experiencing problems that are not evident from looking at the log file?

 

[pirx73]

I only see these errors when router boots. Apart from that i could not find anything in the logs.

 

[ColinTaylor]

I only see these errors when router boots. Apart from that i could not find anything in the logs.

There is nothing in the logs you provided to suggest there was any problem other than on 31st August, even when you have rebooted. So have you had any problems since the 31st August?

The log doesn't say what caused the problem on Aug 31 21:46:37 - my point was that the error messages are old, not re-occurring after boot.

are you still experiencing problems that are not evident from looking at the log file?

 

[pirx73]

@ColinTaylor Wifi devices occassionally drop to 1Mbps speed. Remedy is to disconnect them from wifi and connect back. No error messages in the log when that happens.
Both bands are on fixed channel. All outside wifi networks are very weak - below -80dBm on 2.4GHz, 5GHz band is clean.

 

[ColinTaylor]

@ColinTaylor Wifi devices occassionally drop to 1Mbps speed. Remedy is to disconnect them from wifi and connect back. No error messages in the log when that happens.
Both bands are on fixed channel. All outside wifi networks are very weak - below -80dBm on 2.4GHz, 5GHz band is clean.

This is the first time you have mentioned Wi-Fi and seems unrelated to anything previously discussed in this thread.

 

[pirx73]

You asked me "are you still experiencing problems that are not evident from looking at the log file?"
This is the answer.