RFT: 3.0.0-linode35 and 3.0.0-x86_64-linode20

Linode Staff

I pushed out 3.0.0 kernels yesterday afternoon - it would be great if they could get some testing before pointing Latest to them!

Looks like kernel reserved memory is back down to saner levels for 64 bit (32 bit was corrected in 2.6.39.1).

Thanks!

-Chris

15 Replies

@caker:

Looks like kernel reserved memory is back down to saner levels for 64 bit

…and a great cheer rose up throughout the land!

James

@caker:

Looks like kernel reserved memory is back down to saner levels for 64 bit

Thanks!

-Chris

:twisted:

Seems to be working fine thus far on my ubuntu 10.04 node & a debian 6 node as well though debian one isn't doing much of anything besides nginx currently

It does appear to solve the reserved memory problem introduced in 2.6.39.

However, I did notice that the 3.0.0-x86_64-linode20 kernel appears to somehow break the console (Lish) on my Debian squeeze (6) machine:

The end of the output looks like this (no login prompt):

Loading kernel modules...done.                                                  
Activating lvm and md swap...done.                                              
Checking file systems...fsck from util-linux-ng 2.17.2                          
done.                                                                           
Mounting local filesystems...done.                                              
Activating swapfile swap...done.                                                
Cleaning up temporary files....                                                 
Configuring network interfaces...done.                                          
Cleaning up temporary files....                                                 
Setting kernel variables ...done.                                               
Skip starting firewall: ufw (not enabled)...done.                               
INIT: Entering runlevel: 2                                                      
Using makefile-style concurrent boot in runlevel 2\.                             
Starting enhanced syslogd: rsyslogd.                                            
Starting domain name service...: bind9\.                                         
Starting periodic command scheduler: cron.                                      
Starting dictionary server: dictd.                                              
Starting nginx: nginx.                                                          
Starting NTP server: ntpd.                                                      
Starting Postfix Mail Transport Agent: postfixsshd (1720): /proc/1720/oom_adj is deprecated, please use /proc/1720/oom_score_adj instead.                       
Starting OpenBSD Secure Shell server: sshd.                                     
.                                                                               

Reverting back to Latest paravirt (2.6.39.1-x86_64-linode19) fixes the console again.

I have not looked into this further, I suppose it's something in Debian that breaks with a 3.0 kernel…

A fix for this was deployed this afternoon. Give it another shot.

-Chris

@caker:

A fix for this was deployed this afternoon. Give it another shot.

-Chris

Great, seems to work fine now!

is there someone who booted this kernel on CentOS 6?

Is it ok?

I am giving this a go now.

On my Linode 768 it's showing 740mb of total ram - which means the kernel hogging has reduced (a good thing) - Everything seems to be working well.

thanks for the answer michael… are you using CentOS 6?

It is working well for me.

James

@zunzun:

It is working well for me.

James

Don't you think that an interesting feedback like this could be more interesting if you tell us also what distro are you running?

@sblantipodi:

Don't you think that an interesting feedback like this could be more interesting if you tell us also what distro are you running?

No.

James

Seems to run fine on my Ubuntu 11.04 64-bit.

Working for me so far on Ubuntu 10.04 64 bit.

My total memory went down from:

MemTotal: 502008 kB

to

MemTotal: 500452 kB

It's not a significant difference anyway.

I don't know if this is my imagination, but when testing this kernel I have a feeling that something is kind of off in terms of IO performance/responsiveness.

I'll see if I can gather some actual numbers to support this (or show me wrong), but has anyone else experienced something like this?

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Community Code of Conduct