mailspring
[6883:0414/133339.282540:FATAL:setuid_sandbox_host.cc(157)] The SUID sandbox helper binary was found, but is not configured correctly. Rather than run without sandboxing I’m aborting now. You need to make sure that /usr/share/mailspring/chrome-sandbox is owned by root and has mode 4755.
`trap’ para punto de parada/seguimiento
I am also unsure if the deb-building process is open source? I wasn’t able to find it on github on my own. If someone can link me that I can have a look into the issue and submit a PR.
Hey folks! Hmm, this was an issue with the snapcraft build that we fixed during testing, but afaik it should work fine when Mailspring is installed via the debian package - will take a look at this today. I think that the chrome-sandbox file @bennetgallein mentioned should do the trick, but hopefully we can reelase a patch today or tomorrow with a better fix. I don’t think we want to disable sandboxing except in the snapcraft build (which runs inside the snapd sandbox).
When I install the deb package with sudo dpkg -i <package.deb> it seems to work fine, is there another method of installing the package that is failing maybe? I think that the permissions on this file are important in this release, but weren’t in the past, because we updated Chromium and the new version does some system level sandboxing.
That might be true. I just installed the 1.9.1 update after I applied the fix i mentioned before and ran again in the same issue.
So there seems to be something that messes up the permissions. I also installed with dpkg as usual.
I also noticed that the permission on the file itself (in the .deb archive) are not 4755 but 0755, but I am unsure how that disables chrome-sandbox from running correctly.
I will clone the repository and try if I can find something.
Also you mentioned that it works on your end? I removed the /usr/share/mailspring folder and installed a freshly compiled version from master but without any success really…
But it just seems to be the setuid bit itself… Also saw the fix you implemented yesterday, but as mentioned above it does not seem to effect the binary in the package.