Out Of memory when upgrading to 2.4.2

Hi i have an ePMP as AP , i was able to upgrade al SM connected to that AP to 2.4.2 (from 2.4.1)

 but when i tried to upgrade the AP module an error is shown in web page notification icon:

Error. Unpack failed - damaged or wrong package format.

Looking forward seems to be a problem with memory  logs sent to syslog servers and shows this:

Jun 13 12:44:50 TR1-CB-APW1 kernel snmpget invoked oom-killer: gfp_mask=0x200da, order=0, oom_adj=0
Jun 13 12:44:50 TR1-CB-APW1 kernel Call Trace:
Jun 13 12:44:50 TR1-CB-APW1 kernel [<802492b4>] dump_stack+0x8/0x34
Jun 13 12:44:50 TR1-CB-APW1 kernel [<800aedec>] oom_kill_process+0x6c/0x200
Jun 13 12:44:50 TR1-CB-APW1 kernel [<800af404>] __out_of_memory+0x164/0x198
Jun 13 12:44:50 TR1-CB-APW1 kernel [<800af4ac>] out_of_memory+0x74/0xa0
Jun 13 12:44:50 TR1-CB-APW1 kernel [<800b2594>] __alloc_pages_nodemask+0x488/0x588
Jun 13 12:44:50 TR1-CB-APW1 kernel [<800c2e2c>] handle_mm_fault+0x230/0x7a4
Jun 13 12:44:50 TR1-CB-APW1 kernel [<8006f190>] do_page_fault+0x100/0x2f0
Jun 13 12:44:50 TR1-CB-APW1 kernel [<80065160>] ret_from_exception+0x0/0xc
Jun 13 12:44:50 TR1-CB-APW1 kernel
Jun 13 12:44:50 TR1-CB-APW1 kernel Normal per-cpu:
Jun 13 12:44:50 TR1-CB-APW1 kernel CPU    0: hi:   18, btch:   3 usd:  15
Jun 13 12:44:50 TR1-CB-APW1 kernel active_anon:2578 inactive_anon:1930 isolated_anon:0
Jun 13 12:44:50 TR1-CB-APW1 kernel active_file:0 inactive_file:0 isolated_file:0
Jun 13 12:44:50 TR1-CB-APW1 kernel unevictable:7173 dirty:0 writeback:0 unstable:0
Jun 13 12:44:50 TR1-CB-APW1 kernel free:309 slab_reclaimable:247 slab_unreclaimable:1264
Jun 13 12:44:50 TR1-CB-APW1 kernel mapped:938 shmem:3323 pagetables:64 bounce:0
Jun 13 12:44:50 TR1-CB-APW1 kernel Normal free:1236kB min:1016kB low:1268kB high:1524kB active_anon:10312kB inactive_anon:7720kB active_file:0kB inactive_file:0kB unevictable:28692kB isolated(anon):0kB isolated(file):0kB present:65024kB mlocked:0kB dirty:0kB writeback:0kB mapped:3752kB shmem:13292kB slab_reclaimable:988kB slab_unreclaimable:5056kB kernel_stack:304kB pagetables:256kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
Jun 13 12:44:50 TR1-CB-APW1 kernel lowmem_reserve[]: 0 0
Jun 13 12:44:50 TR1-CB-APW1 kernel Normal: 35*4kB 27*8kB 7*16kB 0*32kB 8*64kB 2*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1236kB
Jun 13 12:44:50 TR1-CB-APW1 kernel 10496 total pagecache pages
Jun 13 12:44:50 TR1-CB-APW1 kernel 0 pages in swap cache
Jun 13 12:44:50 TR1-CB-APW1 kernel Swap cache stats: add 0, delete 0, find 0/0
Jun 13 12:44:50 TR1-CB-APW1 kernel Free swap  = 0kB
Jun 13 12:44:50 TR1-CB-APW1 kernel Total swap = 0kB
Jun 13 12:44:50 TR1-CB-APW1 kernel Out of memory: kill process 8463 (snmpd) score 96 or a child
Jun 13 12:44:50 TR1-CB-APW1 kernel Killed process 8464 (tar)

I use cacti to monitor that AP, and it seems that snmp is doing bad stuff inside

I thougt that one reboot whould do the memory freeing possible but, it's completely necessary? i mean, can i do something so when upgrading reboot once?

Hi, 

You can try performing the upgrade again without rebooting. We've done memory optimizations in release 2.4.3 (currently in beta) and you should not run into this problem. But for now, try upgrading again without a reboot. If not, you may have to perform a reboot. 

Thanks,

Sriram