| 1 |
Andrew Savchenko posted on Sun, 24 Jul 2016 00:30:39 +0300 as excerpted: |
| 2 |
|
| 3 |
> Do we ever had such case like multiple versions of the same |
| 4 |
> single-slotted package installed or recorded as installed in the real |
| 5 |
> world? I'm not sure even in this, but I may assume that it may happen |
| 6 |
> one day. |
| 7 |
> |
| 8 |
> Do we have any proof that PM can recover from such situation, |
| 9 |
> where VDB is a mess and installed files can also be a mess? I'm not sure |
| 10 |
> in this at all. |
| 11 |
> |
| 12 |
> Do we have any test suits for portage (as the most popular PM |
| 13 |
> implementation) for such cases? I doubt this, I can find none. I'm not |
| 14 |
> sure if such tests are implemented in other PM test suits too. |
| 15 |
|
| 16 |
Think of how a package is upgraded (by portage, I don't know enough about |
| 17 |
the others to describe the process for them). The package is built, then |
| 18 |
installed to a temporary location, then "qmerged" from the temporary |
| 19 |
location to the live filesystem, replacing the previous version's files |
| 20 |
and recording the new one in the installed-package database, then the old |
| 21 |
version is unmerged and its record removed from the installed-package |
| 22 |
database. |
| 23 |
|
| 24 |
What happens if there's a crash in either the qmerge or old-version |
| 25 |
unmerge steps? |
| 26 |
|
| 27 |
Right, now there's parts of two versions in the installed-package |
| 28 |
database, and who knows what files from each on the live filesystem. |
| 29 |
|
| 30 |
I'm not a portage dev so won't comment on the test suite angle, but |
| 31 |
portage has been able to handle this with the user simply redoing the |
| 32 |
upgrade (whether from binpkg or full rebuild) for many years now (singe |
| 33 |
before I became a gentooer in 2004, I know as I had some faulty hardware |
| 34 |
at the time and regularly crashed during build and installs, which was |
| 35 |
likely before REPLACING_VERSIONS was a thing), and given the number of |
| 36 |
installations out there and the stress of parallel-building some packages |
| 37 |
while others are installing, the code to handle this is GOING to get |
| 38 |
regularly tested. |
| 39 |
|
| 40 |
This needs to continue to work, thus the PMS rules, and ebuilds that are |
| 41 |
unprepared to deal with it aren't going to help. |
| 42 |
|
| 43 |
-- |
| 44 |
Duncan - List replies preferred. No HTML msgs. |
| 45 |
"Every nonfree program has a lord, a master -- |
| 46 |
and if you use the program, he is your master." Richard Stallman |
Archives