Re: [arch-dev-public] WARNING: changes to screen in testing

2012-10-31 Thread Thomas Bächler
Am 22.10.2012 20:00, schrieb Thomas Bächler: > When you run with systemd's default tmpfiles configuration and you use > screen, your screen sockets are deleted after 10 days. This is bad and > caused me some trouble. > > I changed screen in testing to use /run/screens instead, which won't be > cle

Re: [arch-dev-public] WARNING: changes to screen in testing

2012-10-22 Thread keenerd
On 10/22/12, Thomas Bächler wrote: > However, when upgrading the screen package, your existing screens will > become unaccessible. How should we handle this? Something similar happened on a debian system I administer. Their solution was to have a special temporary binary for connecting to the ol

Re: [arch-dev-public] WARNING: changes to screen in testing

2012-10-22 Thread Thomas Bächler
Am 22.10.2012 20:13, schrieb Dave Reisner: > On Mon, Oct 22, 2012 at 08:00:18PM +0200, Thomas Bächler wrote: >> When you run with systemd's default tmpfiles configuration and you use >> screen, your screen sockets are deleted after 10 days. This is bad and >> caused me some trouble. >> >> I changed

Re: [arch-dev-public] WARNING: changes to screen in testing

2012-10-22 Thread Dave Reisner
On Mon, Oct 22, 2012 at 08:00:18PM +0200, Thomas Bächler wrote: > When you run with systemd's default tmpfiles configuration and you use > screen, your screen sockets are deleted after 10 days. This is bad and > caused me some trouble. > > I changed screen in testing to use /run/screens instead, w

[arch-dev-public] WARNING: changes to screen in testing

2012-10-22 Thread Thomas Bächler
When you run with systemd's default tmpfiles configuration and you use screen, your screen sockets are deleted after 10 days. This is bad and caused me some trouble. I changed screen in testing to use /run/screens instead, which won't be cleared by tmpfiles - and it is the correct place for this s