FC4 test 1 : Out of Memory: Killed process 30798 (named)

[mnisay]

hello .

from FC4 test 1, i can see from dmesg output below ..
my curiosity is with the last output line saying that
"out of memory and have killed the named process by itself? "

my named resolves fine though. CPU load and process are fine.
any hints please.

thanks in advance.
milver nisay

Active:84893 inactive:1273 dirty:0 writeback:0 unstable:0 free:1382 slab:39465 mapped:84977 pagetables:718
DMA free:2088kB min:88kB low:108kB high:132kB active:4564kB inactive:48kB present:16384kB pages_scanned:143 all_unreclaimable? no
lowmem_reserve[]: 0 495 495
Normal free:3440kB min:2804kB low:3504kB high:4204kB active:335008kB inactive:5044kB present:507884kB pages_scanned:10644 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
DMA: 6*4kB 0*8kB 1*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 1*2048kB 0*4096kB = 2088kB
Normal: 158*4kB 9*8kB 29*16kB 9*32kB 1*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 3440kB
HighMem: empty
Swap cache: add 0, delete 0, find 0/0, race 0+0
Free swap = 0kB
Total swap = 0kB
Free swap: 0kB
131067 pages of RAM
0 pages of HIGHMEM
2353 reserved pages
45250 pages shared
0 pages swap cached
Out of Memory: Killed process 22043 (named).

[AndyGreen]

Are you using your local bind to resolve dns?

cat /etc/resolv.conf

Is named restarted somehow?

service named status

Much more interesting question: which process is using all your memory?

[mnisay]

Thanks for the fast reply.

Yes I am using my localhost IP to resolve DNS as I have authoritative DNS.
Yes, named service is up and running, resolving just fine.
Also named does shows that it is tooking a lot of load, its load is normal, my
server load is normal as well, 0.1 to 0.3 load.
Based from my investigations, there no big process that needs high loads.

result of my resolv.conf

search domain.com
nameserver 127.0.0.1
nameserver 213.255.xxx.x
nameserver 213.255.xxx.x

result of service named status:

number of zones: 10
debug level: 0
xfers running: 0
xfers deferred: 0
soa queries in progress: 0
query logging is ON
recursive clients: 4/1000
tcp clients: 0/100
server is up and running


I got the result "Killed Named" from executing #dmesg, now I have to clear the dmesg like so #dmesg -c and monitor it again as I am evaluation FC4 test 1 though.

maybe a FC4 test 1 bug? ..clueless, coz I cant find it with FC1,FC2,FC3 and RH9 logs as well. :(

cheers.

[AndyGreen]

Well, it seems something ate all your memory and so unleashed the OOM killer from its cage. I assume it picked named at random, although it is pretty queer that named is still apparently up. What does

free

say about the state of your memory?

[mnisay]

that was fast. #free results below:

total used free shared buffers cached
Mem: 514856 499684 15172 0 3112 47184
-/+ buffers/cache: 449388 65468
Swap: 0 0 0

my guess seems that i need memory upgrade. however,
weird as it seems, all my other server has all less than 100K free amount of memory regardless of up major services though.

#top command shows

09:36:48 up 193 days, 20:06, 2 users, load average: 0.01, 0.14, 0.25
764 processes: 763 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 1.3% user 1.3% system 0.0% nice 0.0% iowait 97.2% idle
Mem: 1030932k av, 939912k used, 91020k free, 0k shrd, 204224k buff
520004k actv, 23968k in_d, 1952k in_c
Swap: 2048136k av, 176848k used, 1871288k free 235132k cached

CPU load seems to be normal for me. i checked cron logs, nothing seems malicious.
by summary, I am just curious about it as it does not bothers me at all,
moreoever, im just being paranoid only ,and conclusively as well as my uptime has never been bothered by this so far.

thanks so much for you advise sir.

cheers.
milver nisay

[Jman]

test 3 is out. I advise an upgrade as many things have changed since then.

[mnisay]

thanks forall the replies.

thread closed.