| 1 |
On Wednesday, 17 June 2020 10:38:54 BST Neil Bothwick wrote: |
| 2 |
> On Wed, 17 Jun 2020 11:03:00 +0200, Andreas Fink wrote: |
| 3 |
> > > > Thanks for the heads up. To play safe, I emerge passwdqc before |
| 4 |
> > > > emerging @world. It actually emerge without complaint on the three |
| 5 |
> > > > systems I tried. I also made sure I could SSH in after updating PAM |
| 6 |
> > > > before terminating the existing session. |
| 7 |
> > > |
| 8 |
> > > MAKEOPTS="-j1" emerge -1 passwdqc worked here. Sys-libs/pam had |
| 9 |
> > > already installed by the time passwdqc failed the first time. |
| 10 |
> > > Sys-auth/pam is not in my update list, but sys-auth/pambase is, and |
| 11 |
> > > has installed without protest. |
| 12 |
> > |
| 13 |
> > Either USE="-passwdqc" for pambase or MAKEOPTS="-j1" for passwdqc works |
| 14 |
> > consistently for me too. I think you just need to make sure that |
| 15 |
> > pam+pambase are both updated before logging out ;) |
| 16 |
> > One system that I had to rescue from a live system + chroot :( |
| 17 |
> |
| 18 |
> Interestingly, I've now emerged passwordqc successfully on four systems, |
| 19 |
> with MAKEOPTS ranging from -j2 to j8. Only the 4th failed, and that |
| 20 |
> worked when I went from -j4 to -j3. This appears to be a bit of a |
| 21 |
> borderline case, which may explain why it slipped through. |
| 22 |
|
| 23 |
And today, pam and pambase are restored to their earlier versions because sddm |
| 24 |
requires them. But sddm has not changed versions, so why did portage want to |
| 25 |
upgrade pam and pambase on Wednesday? |
| 26 |
|
| 27 |
-- |
| 28 |
Regards, |
| 29 |
Peter. |
Archives