CentOS 5 :: KVM Guest Performance After Kernel/system Update?
Mar 30, 2011
Last I upgraded one of our cent 5 boxes which is running kvm with a mix of centos and windows VMs.After the update performance just tanked on the XP guest, but the others (including other windows VMs) are just fine.I checked with our good friend google and didn't really get any help there.We're hoping to NOT have to rebuild the XP guest, so I thought I'd check in here first.
I am an old days RH release user(from 6.x) and just switching back from Debian/Ubuntu to CentOS on some servers, but I can not understand the kernel update strategy currently enabled in CentOS.There are two boxes, with almost identical installation, but recently there was an auto update of kernel on one box. This auto update also seems to issue an auto reboot on the machine, which is unacceptable on server machines.
I'm experiencing some rather severe problems after updating my Centos 5.4 system (Virtual Xen guest). What happens is that when the system boots it complains about missing .so files which prevents about 50% of the services installed from running. I'm suspecting that it has something to do with selinux for two reasons: 1. The first services to go down complains about the security context of some files, and 2. selinux was kinda the reason I decided to update in the first place as it was disabled when it shouldn't be (enabled in system-config-securitylevel, disabled when running sestatus). The whole boot-sequence ends with alot of "INIT: Id 'X' respawning too fast: disabled for 5 minutes"-messages (including all runlevels) before it goes stale, and I can do nothing. The server in question had undergone very little tinkering from my part, pretty much none at all, the only services installed after installation was apache, mysql and webmin.
Details: CentOS 5.4 is installed on both the host and guest. The guest runs on an lvm-partition. I have two other vm's(also CentOS) running just fine, altough I'm a bit weary of updating them .
Attached are some screenshots of the boot-process.
I hope some one here can share some insight on this problem. It's making me pretty nervous seeing that our whole network is run by CentOS-installations (not that I'm certain that CentOS is the culprit).
My problem is next: I have Public IP-addresses (ex. 80.237.x.x) and this IP-addresses join to network interfaces (eth0:0, ... eth0:5) and I have local IP-address of Guest System 192.168.122.111 of NETWORK 192.168.122.0.What I can do here if I want assign Public IP-address for Guest System? Because if I want connect to my Guest System through any computer in world (ex. through SSH) I should have public IP of guest system (not private, ex.192.168.122.111).
I was using centos for my business applications and now I am trying to work only with opensuse and install my other oprerating systems in it. I was always using vmware , but I decided to try another virtualization technologies other than vmware for testing , I searched the internet and found many other like virtualbox , kvm , xen. I concluded from my search that xen and kvm will be the faster type , I decided to test them, I choose xen, it is better than kvm. I installed opensuse 11.4 and installed xen hypervisor deployed two VMs windows xp and centos 4.8 , they are runing quite good but I have some questions:
1 : Isn't there anyway to improve graphics performance in xen guest , or change the video card memory or type ? 2 : Isn't there any way to copy and paste between the host and guest ? 3 : Isn't there any free application like vmware tools or virtualbox guest tools for xen ? 4 : I use these VMs to install some applications for my geophysics work which requires good graphic performance in the vm , also I don't them to be sluggish sometimes , which is better for that vmware or xen ?
after update to kernel-2.6.18-164.el5 one of the 2 NIC's of my machine are only found at 1 of 4 reboots. Using the old one kernel-2.6.18-128.7.1.el5 all is fine. This are the to NIC's:
The server runs# uname -r2.6.18-128.4.1.el5However, today I executed yum update kernel*due to security advisory. I was just about to reboot the system when I realized that it runs VMWare Server Instance that will most likely fail to restart after kernel upgrade (I had a hard time fixing it after previous kernel update). Now I want to keep 2.6.18-128.4.1.el5 after reboot.I see that new kernel is scheduled for booting:
I have a error when i want to update my system via YUMI execute the next command.sh@ yum -y updateYou could try using --skip-broken to work around the problemYou could try running:package-cleanup --problems package-cleanup --dupes rpm -Va --nofiles --nodigestThe program package-cleanup is found in the yum-utils package.
I have a weird performance issue with a centos 5 running a nfs server and a rh8 client. I think the fact that it is rh8 client should be downplayed. It is just that with rh8 client the performance degradation seems more clear. See test details below OS in server is Centos 5 x86_64 kernel 2.6.18-92.1.22.el5
1Gb connection between machines File to test over NFS is a 1GB file. First of all I wanted to measure how the network alone performs while using NFS. So in the server side I run a "cat" command on the 1GB file to /dev/null. Please note that the disk read speed is about 98MBs. At this point the file system has the 1GB file cached in memory. In the client side a "cat" on the same file gives me a speed of about 113MBs. It seems then that the bottleneck in this instance is the network and it is very close to nominal speed. So the network performance is really good. (BTW I know that the server got that file from cache because a vmstat or iostat shows no disk activity.)
The second test is reading from disk with no caching involve. In the server I flushed the 1GB file from the memory. For instance by reading another 5GB file and I repeat the same thing as above in the client (a cat on the 1GB file). Now, the server has to go to disk.(vmstat or iostat shows the disk activity). However the performance, now, is about 20MBs, I was expecting something closer so 90MBs. (since the reading speed in the server in the first test showed 98MBs).
This second test was repeated for ext2, ext3, xfs with no significant differences. A similar test using a RH8 NFS server and client gets me close to 60MBs for a 1GB file not cache by the file system in the serverSince network speeds and disk read speeds are not the bottlenecks ... what or where is the limiting factor then?
I have centos 5.3 workin on mini itx atom 330. i have some problem with the network when i use Samba. when i move big files the network goes down. i wanna to install the new kernel on my centos to try to fix the network problem with the new drivers includes on the new kernel.
I applied the Kernel update that showed in the updater via the GUI.It now won't boot anymore.Memory for crash kernel (0x0 to 0x0) notwithin permissible range.WARNING calibrate_APIC_clock: the APIC timer calibration may be wrong.
when a remove a user from the system using the domain userdel guest the user gets removed but the /home/guest doesn't get removed so i remove it by command rm -rf /home/guest then i recreate the user by giving in the command useradd guest now it gives the error mailbox file already exists what does it mean when it says so though this command creates the user.
i upgraded my kernel to the latest version that being linux-image-2.6.31-19 and now when i start my linux box i get the grub shell. this problem happend once before but i only had to run the command: sudo update-grub2 and everything turned back normal. now even though i run the command everytime i restart my linux i get the grub shell.
Firstly my gdesklets have gone, and if I try to run gdesklets then I get a blank window in the top left corner, followed by another blank window in the centre which disappears, then the first one fades to gray.
Secondly, clicking the 'running man' to exit just closes both gnome panels.
Ubuntu 8.04 (2.6.24-25)I know 8.04 is pretty old now, and I should probably start over, but it was all fine until this latest update.
i'm currently running centos 5 with kernel version 2.6.18-92.1.22.el5 is there any way to update to kernel version >= 2.6.27 via yum ? if not via yum is it advisable to compile from source?
I try to update my kernel to 2.6.18.128.4.1.el5 using # yum update
After # yum update # rpm -qa kernel*|sort gives the following: kernel-2.6.18-128.1.14.el5 kernel-2.6.18-128.1.16.el5.centos.plus kernel-2.6.18-128.4.1.el5.centos.plus kernel-2.6.18-92.1.18.el5 kernel-2.6.18-92.el5 kernel-devel-2.6.18-128.1.14.el5 .....
This my grub.conf: # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hda default=1 timeout=5.....
yum update installed: CentOS (2.6.18-128.4.1.el5.centos.plusxen) and CentOS (2.6.18-128.4.1.el5.centos.plus)
both are not working. I have to run the system using: 2.6.18-128.1.14.el5. Upon trying again #yum update it says that no packages are marked for update. Can I manually install the desired kernel? Do you know a reason why yum update is not automatically installing the proper kernel?
I had a Centos 5.5 system working great...and then due to crazy curiosity, installed the kernel update. Did this through the graphical interface, uncheck all update other than kernel. It downloaded, installed fine..After reboot now all I get is GRUB_ and then nothing else! I waited 3-4 minutes and had to shut down system. Have been googling around but thought I will post it here as well..When I boot with Win7 and use Ext2fsd to check, the boot folder is empty. Not sure if this is because ext2fsd does not read boot folder or it got wiped out by the yum upgrade process.
I have recently run an update and now we are unable to boot the newer kernel. We get stuck as below;
Memory for crash kernel (0x0 to 0x0) notwithin permissible range Red Hat nash version 5.1.19.6 starting Unable to access resume device (/dev/VolGroup00/LogVol06)
[Code]...
The kernel version we were at was 2.6.18-194.el5 and the update move us to 2.6.18-194.17.1.el5. I've done some googleing but I'm unable to work out exactly what I need to do. I'm guessing the new kernal is missing a module?
I have a problem with the kernel update to 2.6.34.6. Up until 2.6.33.x my system boots fine, but with this update the boot stops at the moment that the mouse cursor should become visible. To resolve the problem, I've gone back to 2.6.33.x and removed the 2.6.34 kernel but I wonder what happens with the next kernel update. Anyone else having this problem?
It's taken a lot of work, but I have figured out over thecourse of many many fresh installs over the past few days that the kernel updatefrom the update manager in ununtu is breaking my system. This is both for the -22 and -27 updates.Is anyone else having problems with these and/or is there a fix?
I am running CentOS 5.x and want to yum update, but I do not want to touch the kernel. How can I exclude the kernel update and prevent yum from removing my current kernel?
I'm using Cpanel on my server, before upgrade to new kernel, i could see 8 cores on my i7 server, now i only see 4 cores, is there anything that i need to do to have another time the 8 cores?
I have 2.6.18-164.2.1.e15 running on Hyper-v. I did yum update and noticed that kernel is upgraded to 2.6.18-164.9.1.e15. If I try to boot into the new kernel it hangs with message first "switchroot mount failed" and then "kernel panic". This is probably related to Hyper-v drivers installed in the old kernel. First question: Do you think this is correct? Is it possible to somehow disable Hyper-v drivers so I can boot into the new kernel and then reinstall drivers? There is nothing in Microsoft documentation about uninstalling Linux Integration Components.
I installed the Centos 5.5 with XEN and put 03 virtual machines (Centos 5.5 too). I did this correctly, but I have a problem. In the virtual machine, I need update the kernel. I used "yum update kernel" and this is downloaded correctly and the menu.lst file is updated too, but the kernel not ran. I ran "grub-install /dev/sda" or "grub-install /dev/xvda" and received a error return. I read the device.map file and i ran "grub-install /dev/sde" and received a error return too.
The errors: [root@vm01 ~]# grub-install /dev/sda /dev/sda: Not found or not a block device. [root@vm01 ~]# grub-install /dev/sde /dev/sda1 does not have any corresponding BIOS drive. [root@vm01 ~]# grub-install /dev/xvda /dev/xvda: Not found or not a block device.
I am currently working to get to a stable dual boot Win7/CentOS 5.5 configuration on a Dell Latitude E6510 with the Intel 82577LM chipset. I haven't gotten the Centrino N WiFI chipset working yet, but the Eth0 device was working just fine.However, yesterday I did a system update. I did not pay close attention, but I believe the update did mention networking, because I thought it might help me get my wireless going. Now, Eth0 fails with the message, "Determining IP information for eth0... failed; no link present. Check cable?" It does this at boot up and if we try reactivating it once the system is up. We have tried unstalling and re-installing the driver.The Window 7 boot sees the adapter and runs just fine. I was at a different site location yesterday when everything was working, but my co-workers and I can't help but think it might be related to the update yesterday, since everything worked prior to the update and Windoze 7 works fine with the adapter.
I have not been able to mount my NTFS harddrive on the CentOS side, despite installing fuse, and I don't have wireless, which makes it difficult to get access to new drivers. I also have a FAT32 partition that would be great to share between the boots for passing info, but I haven't found a way to mount a FAT32 partition in CentOS.
