Percona Toolkit 2.2.2 released; bug fixes include pt-heartbeat & pt-archiver

During the Percona Live MySQL Conference & Expo 2013 the week before last, we quietly released Percona Toolkit 2.2.2 with a few bug fixes:

  • pt-archiver –bulk-insert may corrupt data
  • pt-heartbeat –utc –check always returns 0
  • pt-query-digest 2.2 prints unwanted debug info on tcpdump parsing errors
  • pt-query-digest 2.2 prints too many string values
  • Some tools don’t have –help or –version
  • Some tools use @@hostname without /*!50038*/
  • pt-stalk prints the wrong variable name in verbose mode when –function is used
  • And more

Percona ToolkitThe first two bugs are important. Those using pt-archiver –bulk-insert with UTF-8 data should definitely upgrade. Those using pt-heartbeat –utc should also definitely upgrade.

Users may note the revival of the –show-all option in pt-query-digest. This had been removed in 2.2.1, but resulted in too much output in certain cases.

A new –recursion-method was added to pt-table-checksum: cluster. This method attempts to auto-discover cluster nodes, alleviating the need to specify cluster node DSNs in a DSN table (–recursion-method=dsn).

See for the full list of bugs fixed in Percona Toolkit 2.2.2.

Visit to download Percona Toolkit 2.2.2; it’s also in our main repos.

Share this post

Comments (2)

  • Evelio Vila

    currently using 2.2.2 and
    mysql Ver 14.14 Distrib 5.5.30, for Linux (x86_64) using readline 5.1
    i am getting this since the upgrade of percona this week:

    pt-heartbeat -F/etc/my.cnf -Dpt_heartbeat –check

    The –master-server-id option must be specified because the heartbeat table pt_heartbeat.heartbeat uses the server_id column for –update or –check but the server’s master could not be automatically determined.
    Please read the DESCRIPTION section of the pt-heartbeat POD.

    any ideas why this could be happening?
    how does the master’s auto detection takes place?


    don’t know what went wrong since i upgraded

    May 15, 2013 at 11:09 am
  • Evelio Vila

    Actually PTDEBUG=1 did the trick 😉 it was an encoding issue on the table.

    May 16, 2013 at 4:35 am

Comments are closed.

Use Percona's Technical Forum to ask any follow-up questions on this blog topic.