Status update: Proxmox VE 2.6.32 with OpenVZ - still not ready

[martin]

Hi all,

The last weeks we did a lot of testing with the latest OpenVZ on 2.6.32 but there are still some issues, especially KSM is not working and also fsync/sec are much too slow for a production server (all these issue seems to be related to the OpenVZ patches)

We will monitor these issues and keep you informed if there are any news.

Martin

 

[chronos]

These issues are mostly related to quality of full virtualization I guess. I have running 2.6.32 kernel on 5 servers used for OpenVZ hosting and they are running good enough. I haven't any running KVM virtual machine but recently I tested one and there was issue with disk performance. Much worst than on early tested 2.6.24 pve kernel. So if somebody be enough with openvz only hosting 2.6.32 kernel should be good. This could be similar approach as 2.6.32 KVM only kernel.

 

[tom]

can you post your pveperf results from a 2.6.32 openvz kernel? (and include hardware details, raid controller, settings)

 

[chronos]

I think I can't provide enough information to identify source of problem. I have five servers in cluster and all of them are running proxmox 1.5 with 2.6.32 openvz kernel and are mainly used for openvz virtualization. I have only one KVM VM and it is aimed for testing and running remotely some common windows applications.

# pveperf
CPU BOGOMIPS: 11251.23
REGEX/SECOND: 474096
HD SIZE: 94.49 GB (/dev/mapper/pve-root)
BUFFERED READS: 90.50 MB/sec
AVERAGE SEEK TIME: 10.46 ms
FSYNCS/SECOND: 109.33

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 5600+ 2,8 GHz
HDD: WDC WD15EARS-00Z5B1 1,5 TB SATA Green

# lspci
00:00.0 Host bridge: Advanced Micro Devices [AMD] RS780 Host Bridge Alternate
00:01.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (int gfx)
00:0a.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (PCIE port 5)
00:11.0 SATA controller: ATI Technologies Inc SB700/SB800 SATA Controller [AHCI mode]
00:12.0 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI0 Controller
00:12.1 USB Controller: ATI Technologies Inc SB700 USB OHCI1 Controller
00:12.2 USB Controller: ATI Technologies Inc SB700/SB800 USB EHCI Controller
00:13.0 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI0 Controller
00:13.1 USB Controller: ATI Technologies Inc SB700 USB OHCI1 Controller
00:13.2 USB Controller: ATI Technologies Inc SB700/SB800 USB EHCI Controller
00:14.0 SMBus: ATI Technologies Inc SBx00 SMBus Controller (rev 3c)
00:14.1 IDE interface: ATI Technologies Inc SB700/SB800 IDE Controller
00:14.2 Audio device: ATI Technologies Inc SBx00 Azalia (Intel HDA)
00:14.3 ISA bridge: ATI Technologies Inc SB700/SB800 LPC host controller
00:14.4 PCI bridge: ATI Technologies Inc SBx00 PCI to PCI Bridge
00:14.5 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI2 Controller
00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Miscellaneous Control
01:05.0 VGA compatible controller: ATI Technologies Inc RS880 [Radeon HD 4200]
01:05.1 Audio device: ATI Technologies Inc RS880 Audio Device [Radeon HD 4200]
02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 02)
03:0e.0 FireWire (IEEE 1394): Texas Instruments TSB43AB23 IEEE-1394a-2000 Controller (PHY/Link)

Disc controller should work in AHCI mode. Only one disc is present, no RAID mode active.

In openvz container performance is good as could be expected. In KVM VM Windows XP is somehow bad with disc I/O operations and system stop responding even on little disc load. It seems to be rather count of I/O operations or seek problem than raw sequence read/write.

Copy big file inside VM on same disc get about 5-10 MB/s.


I can say OpenVZ on kernel 2.6.32 is ready. At least as testing pve kernel could be welcomed by some users. At least for non critical use.
KVM may need some additional tweaking. KVM only pve kernel is already provided so openvz only kernel can be provided too

 

[dietmar]

In openvz container performance is good as could be expected.

NO, it is not. Please try to run a database/transaction benchmark - or simply 'pveperf' - You will get increadible low fsync rate.

Also, KSM is totally broken.

 

[tux]

There is a new version of OpenVZ for 2.6.32. Check out : http://wiki.openvz.org/Download/kernel/2.6.32/2.6.32-bykovsky.1 .

There is also a new version for 2.6.18 - http://wiki.openvz.org/Download/kernel/rhel5/028stab070.4

Maybe you can add it to pvetest - so we can test it and give feedback to the OpenVZ developers? If this is to complex (or you dont want it) you can write some lines how I can switch to the OpenVZ 2.6.32 and I can spend my time with bughunting / bugreporting. If nobody report bugs to the OpenVZ team, we will never have a stable kernel.

 

[tom]

There is a new version of OpenVZ for 2.6.32. Check out : http://wiki.openvz.org/Download/kernel/2.6.32/2.6.32-bykovsky.1 .

See http://bugzilla.openvz.org/show_bug.cgi?id=1622

There is also a new version for 2.6.18 - http://wiki.openvz.org/Download/kernel/rhel5/028stab070.4

See http://bugzilla.openvz.org/show_bug.cgi?id=1610

Maybe you can add it to pvetest - so we can test it and give feedback to the OpenVZ developers? If this is to complex (or you dont want it) you can write some lines how I can switch to the OpenVZ 2.6.32 and I can spend my time with bughunting / bugreporting. If nobody report bugs to the OpenVZ team, we will never have a stable kernel.

Feel free to add your comments and test results to these bugs.