From OpenVZ Virtuozzo Containers Wiki
Jump to: navigation, search


  • Driver updates
  • Fixed serious bug in conntracks
  • Small security fix
  • Thread exec fix
  • vzquota off optimization
  • Oops decoding is tunable now


Same as 022stab072.2, plus:



  • -CONFIG_SCSI_QLA6322=m

Driver updates

  • Updated DRBD to 0.7.17,
  • DM layer updated to RHEL4u3,
  • tg3 driver update up to 3.43.rh,
  • sk98lin driver update up to version,
  • e1000 driver update up to 6.1.16-k3 version.



Patch from mainstream:

[TCP]: Do not use inet->id of global tcp_socket when sending RST.

The problem is in ip_push_pending_frames(), which uses:

if (!df) {
        __ip_select_ident(iph, &rt->u.dst, 0);
} else {
        iph->id = htons(inet->id++);

instead of ip_select_ident().

Right now I think the code is a nonsense. Most likely, I copied it from old ip_build_xmit(), where it was really special, we had to decide whether to generate unique ID when generating the first (well, the last) fragment.

In ip_push_pending_frames() it does not make sense, it should use plain ip_select_ident() instead.

Signed-off-by: Alexey Kuznetsov <>
Signed-off-by: David S. Miller <>


Patch from Denis:

This patch removes veprintk related warnings in ia64/ia32/sys_ia32.c


Patch from mainstream:

[PATCH] disk barriers: core

This patch adds error code to submit_bh(), which is required by ext3 online resize and dm update.

Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>


Patch from Dmitry:

Fixed oops in NAT due to attempt to cleanup already freed area. Bug #59950.


Patch from Dmitry:

VPS could issue SO_GET_ORIG_DST get sockopt while they have no ip_conntracks enabled. This leads to oops. Added check to return ENOPROTOOPT for such case.
Bug #60794.


Patch from Kirill:

This patch fixes usage of kernel type cycles_t in user interface header. related to OpenVZ Bug #123.


updated drbd driver updated to 0.7.17


[DRIVER UPDATE] Qlogic qla2xxx driver updated to 8.01.03 version (site)

Sources from HP site, prepared by Kostja: Qlogic qla2xxx driver updated to 8.01.03 version.

Obsoleted: diff-ms-gcc4-qla2xxx-20051103

Bug #27641.


Patch from mainstream:

[PATCH] atomic_inc_return() for i386


Patch from mainstream:

[PATCH] atomic_inc_return() for x86_64


Patch from mainstream:

Add kstrdup() function. Required for dm update.


[UPDATE] updated dm layer to RHEL4u3 version.

Required, as RedHat broken compatibility of new user tools with old kernels.

OpenVZ Bug #120.
Bug #60309.


Patch prepared by Vasily:
tg3 driver updated up to 3.43.rh version,
sources were taken from RHEL4U3 2.6.8-32EL kernel

Obsoleted linux-

Bug #15462.


Patch prepared by Vasily:
sk98lin driver was updated up to version,
sources were taken from

Obsoleted linux-

Bug #28918.


Patch from Vasily:

mii driver update, sources were taken from RHEL4U3 2.6.9-34EL kernel, required for tg3 driver


PCI ID updated to RHEL4U3 kernel version (2.4.9-34EL)


Patch from Andrey:

This patch fixes hiddev to make kernel compile on ia64 arch. Caused by linux-2.6.8-dm-20051004.patch


Patch prepared by Vasily:

e1000 driver updated up to 6.1.16-k3 version

sources were taken from RHEL4U3 kernel 2.6.9-34EL

obsoleted linux-

OpenVZ Bug #131.

Bug #19952.


Patch from Kirill:

vzquota off syncs inodes, so that inodes are synced one by one and waited for. This is slow.

This patch changes the logic: all inodes should be kicked for syncing first, and then only waited for. This makes VPS creation to be faster.


Patch from Kirill:

Added sysctl variable to disable automatic call traces decoding. Required for machines not connected to any kind of console.


Patch from Kirill:
Fix of user space headers for vzctl.

Bug #61057.


Patch from Dmitry:

Fixed oops not fully fixed by diff-ve-ct-getorigdst-20060331


Patch from Dmitry:

Fixed oops on iptables proc entries access inside VPS if iptables are not permitted in this VPS.

Bug #60958.


Patch from Vasiliy:

setxattr and getxattr have to check CAP_VE_ADMIN in order to return proper errors inside VE, similar to host.

Bug #59526.


Patch from Pavel:

If a non-leader thread calls exec it becomes a thread group leader and thus SET_VE_LINKS/REMOVE_VE_LINKS will manipulate list on this task. Meanwhile such task after exec will not be in list because it was not added there on fork. This may lead to creation of invisible by ps task, unstopable VPS, since do_initproc_exit will not find this task to kill it, and even ve task list corruption.


Patch from Kirill:
This patch fixes OVZ kernel compilation with some new gcc.