Difference between revisions of "Download/criu/3.4"

From CRIU
Jump to: navigation, search
(Created page with "{{Release|3.4}} <!-- NOTE: you can use {{Bug|123}} to link to a github issue --> === New features === === Bugfixes ===")
 
 
Line 3: Line 3:
  
 
=== New features ===
 
=== New features ===
 +
* Support for s390x architecture
 +
 +
=== Improvements ===
 +
* Unexpected death of restored tasks is reported with more details in logs
 +
* Merged many [[images]] containing info about files into one big <code>files.img</code>
 +
* When helper utility fails (ip, iptables, tar) its name is printed in logs
  
 
=== Bugfixes ===
 
=== Bugfixes ===
 +
* Compilation failed on newer glibcs (ucontext_t)
 +
* Dying helper task could deadlock the restore process
 +
* Install-related makefile variables weren't configurable for distro build
 +
* SIT (ipv6-to-v4 tunnel) presence on host blocked dump of any containers
 +
* Potential NULL dereference when dumping net namespace
 +
* Dump via [[page server]] might not work across different criu versions
 +
* Failure to restore a subtask could be ignored by the restore command
 +
* EOF on page-server socket wasn't handled

Latest revision as of 13:07, 18 August 2017

Tarball: criu-3.4.tar.bz2
Version: 3.4 "Cobalt Swan"
Released: 21 Aug 2017
GIT tag: v3.4

New features[edit]

  • Support for s390x architecture

Improvements[edit]

  • Unexpected death of restored tasks is reported with more details in logs
  • Merged many images containing info about files into one big files.img
  • When helper utility fails (ip, iptables, tar) its name is printed in logs

Bugfixes[edit]

  • Compilation failed on newer glibcs (ucontext_t)
  • Dying helper task could deadlock the restore process
  • Install-related makefile variables weren't configurable for distro build
  • SIT (ipv6-to-v4 tunnel) presence on host blocked dump of any containers
  • Potential NULL dereference when dumping net namespace
  • Dump via page server might not work across different criu versions
  • Failure to restore a subtask could be ignored by the restore command
  • EOF on page-server socket wasn't handled