Memory squeeze,deferring packet.

J

Jordi

Verified User
Joined
Jan 14, 2006
Messages
15
Does anybody knows what this is, my server was suddenly unreachable and nothing was done on it.
Is that a problem caused by DA (did the update to the latest version)
In the datacenter they did a system reboot to solve this?
Was that ok or could there be a other solution?

CentOS 4.0
DA latest version


Apr 18 21:36:25 hosting kernel: eth0: Memory squeeze,deferring packet.
Apr 18 21:36:25 hosting kernel: eth0: NULL pointer encountered in Rx ring, skipping
Apr 18 21:36:56 hosting last message repeated 18307 times
Apr 18 21:37:01 hosting last message repeated 2228 times
Apr 18 21:37:01 hosting crond(pam_unix)[14245]: session opened for user root by (uid=0)
Apr 18 21:37:01 hosting kernel: eth0: NULL pointer encountered in Rx ring, skipping
Apr 18 21:37:02 hosting last message repeated 21 times
Apr 18 21:37:02 hosting crond(pam_unix)[14245]: session closed for user root
Apr 18 21:37:02 hosting kernel: eth0: NULL pointer encountered in Rx ring, skipping
Apr 18 21:37:33 hosting last message repeated 2104 times
Apr 18 21:38:00 hosting last message repeated 409 times
Apr 18 21:38:01 hosting crond(pam_unix)[14248]: session opened for user root by (uid=0)
Apr 18 21:38:01 hosting crond(pam_unix)[14248]: session closed for user root
Apr 18 21:38:01 hosting kernel: eth0: NULL pointer encountered in Rx ring, skipping


thank you
 
It looks like your system has run out of memory. Do you have a swap partition set up? Is it 1.5 times the amount of RAM you have in the system?
 
ldjnetworks,

Where did you get the idea you need 1.5 times memory for your swap file?

Lots of us don;t believe that, though I know I've read it before.

Jordi, what OS distribution?

I found a lot by googling Memory squeeze,deferring packet.

Jeff
 
CentOS 4.0

I also found a lot, to mutch to know what went wrong

thank you
 
JLasman,
That is what I have always been taught. I've taken several classes in order to get my RedHat CE certificate, and all of them have recommended 2.0*RAM.

From RedHat's site:
http://www.redhat.com/docs/manuals/...-multi-install-guide/s1-diskpartitioning.html

Creation of the proper amount of swap space varies depending on a number of factors including the following (in descending order of importance):

*The applications running on the machine.
*The amount of physical RAM is installed on the machine.
*The version of the OS.

Swap should equal 2x physical RAM for up to 2 GB of physical RAM, and then 1x physical RAM for any amount above 2 GB, but never less than 32 MB.

Using this formula, a system with 2 GB of physical RAM would have 4 GB of swap, while one with 3 GB of physical RAM would have 5 GB of swap. Creating a large swap space partition can be especially helpful if you plan to upgrade your RAM at a later time.
Click to expand...
 
Thanks for pointing it out. I always knew I read it somewhere but I never remembered where.

There was a time when Unix servers would do a complete dump to the swap partition when they crashed (if they could) so there was a reason why you should have at least the same as memory.

But Linux has never done that (read further for my source).

The whole concept started back when servers were running a few Megabytes of memory (I've been using Unix since at ran on a 128K machine) and you could easily run out of memory.

Every article on swap memory that gives a reason for large amounts of swap space explains that on a server you need it in case you have runaway processes.

However this idea falls apart when you realize that the server is going to come to a slow crawl... such a slow crawl that you'll end up doing a hardware reboot rather than wait the hours it might take you to get in and fix it, by the time it's using most of even a small swap partition.

So who was it who told me that Linux didn't require any specific amount of swap space for dumping memory, and that it wasn't important to have a lot of swap on a large server? None other than Robert Love, the author of several Kernel books (published by his employer Novell), and one of the well-known kernel contributors.

Is he wrong? Maybe.

We use 1 Gig of memory and 1 Gig of swap partition space on many of ours servers; I'm one of those people who started with Unix back in the 70s, and old habits die hard.

I guess that I and a lot of other admins would flunk those tests. But that's okay.

I've edited my post to sound a little less subjective.

Jeff
 
JLasman,
The certs mean nothing really, just a means to an end (job). I've been using RedHat since 4.2 (still have the CDs :) ) so I haven't been around quite as long as you, but I've always done 1.5X. Old habits die hard I guess. Thanks for the history, that's interesting. I did not know the reason behind it.

Jordi,
This may be your problem. From your PHPSysinfo:

Physical Memory 99% 2.74 MB 437.71 MB 440.46 MB
Click to expand...

You've currently got 2MB free memory, although none of your swap is being used, so I'm not quite sure why you're getting that error.
 
Add some memory to your server Jordi. I am not real sure what exactly causes your error but it seems to be memory related. When they start talking about atomic memory etc etc my eyes tend to glass over. Seems to be a problem but it does seem that adding some memory to your box would help prevent it.

As for swap I agree that having tons of swap is not needed in most cases. I mean you would not want your system using 2GB of swap anyway. And as Jef pointed out if a process sucks all your RAM then starts eating your swap having all that swap will not really help anyway.

LOL here is my "old" habit, 1GB swap as a default. If I see it getting used much I add RAM.
 
Thx guy's for the reply's

I guess i have to invest more in memory ;)

thank you
 
You must log in or register to reply here.
Back
Top