Problem with internal mails, when exim is executed from within a directory that doesnt exist anymore

[iLogge]

Operating System (OS/VERSION):

debian 8

VestaCP Version:

latest, Vesta 0.9.8-17

Installed Software (what you got with the installer):

apache only, but rest is pretty default (exim, dovecot etc)

Steps to Reproduce:

open "v-back-user" shell script file and add:

pwd
cd /backup
pwd

before:

cat $BACKUP/$user.log |$SENDMAIL -s "$subj" $email $notify

then run a backup from the panel and you will see something like:

2016-11-29 15:05:01 Local: /backup/admin.2016-11-29.tar
2016-11-29 15:05:01 Size: 1 Mb
2016-11-29 15:05:01 Runtime: 1 minute
/backup/tmp.XeVtfBjspX
/backup

inside the vestacp backup.log. and since the tmp directory doesnt exist anymore, when sendmail is called, exim fails and doesnt send any mail:

shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory
shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory
shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory
exim: can't get the current working directory: No such file or directory

Related Issues/Forum Threads:

https://forum.vestacp.com/viewtopic.php?f=41&t=13214

Other Notes:

there might be also other occasions, where this error is true, not only for the backup script
(cron, restore, etc..)

[Skamasle]

Is common linux warning if dir not exists then return warning and something not working, maybe good idea add some test to check if dir exists before do something.

[iLogge]

its actually enough to just ad an "cd /backup" in front of every direct sendmail occasion from the console.
i changed it allready on my server, but those edits are not update-proof, as long as vesta not add those checks :)

[dpeca]

Fixed - 930a2a8