Changed ppa repository, updated, upgraded, but launching fails

Using Ubuntu 14.04 LTS. I used to fetch my Libreoffice from the ppa repository and LO 5.4 worked as well as it could do.

I have then updated the repository to, which is not version-specific and proved to work well in another computer. I launched the classical sudo apt-get update and sudo apt-get upgrade to fetch and install the latest LO version.

However, launching LO fails. If I do so from the command line I get the error notification

/usr/lib/libreoffice/program/soffice.bin: symbol lookup error: /usr/lib/libreoffice/program/ undefined symbol: _ZN11LanguageTagC1ERKS_

FYI, libreoffice --version returns “LibreOffice 00m0(Build:2)”, so the upgrade has been successfull in some sense. I also tried to remove LO with sudo apt-get remove libreoffice and reinstalling with the latest settings again. This error is persistent.

What is the cause and fix for this issue? Thanks for helping out

Any help in this AksUbuntu Q&A? Your error notification is included in the (much longer) one reported there.

This problem is specific to your Operating System. Probably it is unrecoverable. LibreOffice uses libraries that do not exist in Ubuntu 14.04 or have conflicts with those already installed. If you want to use LO other than the official supported by your OS, then you definitely should upgrade your OS.

Disagree. LO 6.0 works well in the other computers where Ubuntu 14.04 is installed as well. The problem appears to be specific to this installation procedure. See also alternative answer above, which I discover in the meantime

This is an ephemeral incident. See inline picture, where launching libreoffice fails on two calls on day 1 (instructions #2009, #2010), and proceeds on day 2 (calls #2011). The instructions 2009 and 2010 are those reported upon above. Note that the time of the first call 2011 is incorrect.

image description

All that happened between day 1 and 2 is that I have edited some document with a plain text editor and put the computer in suspended mode. Digression: then the time of the first 2011 call is probably when the computer was put at sleep on day 1, which persisted in the shell clock when waking up on day 2. Also note that non-fatal ‘cannot parse’ errors only show up at the first #2011 call. Afterwards all proceeds seamlessly.

This answer provides no explanation for such a glitch. Waiting a little before making new attempts can be a workaround.