1. 22 Feb, 2021 8 commits
  2. 21 Feb, 2021 3 commits
  3. 18 Feb, 2021 5 commits
    • boyska's avatar
      Merge branch 'test/enable-systemd-run-garbage-collection' into 'stable' · baa481b0
      boyska authored
      Test suite: enable garbage collection of failed transient units created by systemd-run
      
      See merge request tails/tails!342
      baa481b0
    • boyska's avatar
      Merge branch 'test/18103-triage-tests' into 'stable' · b0a33561
      boyska authored
      Test suite: add and document Cucumber tags
      
      See merge request tails/tails!348
      b0a33561
    • Tails translators's avatar
      merge with main git using merge_canonical_changes.py. · 35350fce
      Tails translators authored
      wiki/src/install/inc/steps/pc_boot_menu.inline.ar.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.ca.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.id.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.pl.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.ru.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.sr_Latn.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.tr.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.zh.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      wiki/src/install/inc/steps/pc_boot_menu.inline.zh_TW.po: updating as wiki/src/install/inc/steps/pc_boot_menu.inline.de.po has changed.
      35350fce
    • IkiWiki's avatar
      updated PO files · 1a58b24e
      IkiWiki authored
      1a58b24e
    • sajolida's avatar
      Stop tell people to contact me · 50fc5424
      sajolida authored
      - It's hasn't been that useful in the past.
      - Most people shouldn't need to Boot Menu key anymore.
      50fc5424
  4. 17 Feb, 2021 10 commits
  5. 16 Feb, 2021 4 commits
    • sajolida's avatar
      Send half of our bitcoins directly to Riseup Labs · 56865961
      sajolida authored
      We have enough bitcoins in bitcoind for now.
      56865961
    • sajolida's avatar
      Merge remote-tracking branch 'origin/master' · 2bb3e8b2
      sajolida authored
      2bb3e8b2
    • boyska's avatar
      Merge branch 'test/sysadmin-17605-migrate-test-git-repo' into 'stable' · 1d324ae3
      boyska authored
      Test suite: adjust to new test Git repository location
      
      See merge request tails/tails!344
      1d324ae3
    • intrigeri's avatar
      Test suite: give the Linux kernel's memory poisoning feature time to do its job · c7afdf86
      intrigeri authored
      I've seen "Scenario: Erasure of memory freed by killed userspace processes" fail
      once on one of my fastest tester VMs, with "Pattern coverage: 3.796% (52 MiB out
      of 1372 MiB reference memory)".
      
      Note that we previously filled 128 MiB of RAM with the known pattern,
      and we find less than that here (52 MiB).
      
      This leads me to this hypothesis:
      
      1. When we exit from the "I kill the allocating process" step,
         the allocating process was killed (we know this) and memory poisoning
         has started.
      
      2. When we dump the memory in "I find very few patterns in the guest's memory",
         memory poisoning has not finished its job yet: there's still 52 MiB
         of memory left to clean up.
      
      I.e. there's a race condition between Linux memory poisoning and us dumping the
      memory, which is almost always won by Linux memory poisoning, but in the failed
      run I mentioned, memory poisoning lost the race.
      
      I don't see how we can wait specifically for memory poisoning to complete before
      we dump the memory, so I'm adding a 1s delay here, to increase the chances
      we don't lose this race condition.
      
      refs #18188
      c7afdf86
  6. 15 Feb, 2021 10 commits