Summary: | baselayout uses `mktemp` but doesnt RDEPEND on it | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | t35t0r <t35t0r> |
Component: | [OLD] Core system | Assignee: | Gentoo's Team for Core System packages <base-system> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | ivo.welch, michael, stefan |
Priority: | High | ||
Version: | 2006.1 | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | 169082 | ||
Bug Blocks: | |||
Attachments: | emerge --info |
Description
t35t0r
2007-03-06 19:31:50 UTC
Created attachment 112306 [details]
emerge --info
emerge --info
Had the same issue, but corrected it by emerging mktemp, then running /sbin/depscan.sh I also had the same issue, only for me it was quite serious as my machine wouldn't run any of the init scripts. This meant that I didn't have network access so I couldn't run emerge. I had to download the source on another machine and copy it across, build it locally and run depscan.sh -u. Once I did this, the system booted up fine, and I could emerge mktemp. *** Bug 170926 has been marked as a duplicate of this bug. *** I have also had this problem. Caused serious problems on my desktop (similar to Rick Moynihan). Also saw it on another machine I maintain but managed to spot it and fix it before a reboot. This is a really serious problem (breakage in initscripts!) and only needs someone to add mktemp to RDEPEND. that is not possible ... please review the bugs this one depends on before assuming it's doable at this point in time (In reply to comment #6) > that is not possible ... please review the bugs this one depends on before > assuming it's doable at this point in time > To my opinion that is not true. As long as mktemp will block (as it is masked) the update of baselayout. This is 1000 times better than an unbootable system (without mktemp). Priority and severity of this bug should be set to highest available ! it doesnt matter, nothing is going to change here until the bugs that block this are resolved (In reply to comment #8) > it doesnt matter, nothing is going to change here until the bugs that block > this are resolved > It is not the bug that blocks. It's the missing dependency that turns running systems into unusable ones. Have a look at the following scenario: Version 1 of component A works. Version 2 of component A needs component B to work. Component B does not work on systems (X, Y) but on systems (W, V). If version 2 of A gets installed on system X, such users have no chance to get a working system back. If version 2 of A gets installed on system W, users have to try to get B from somewhere to get a running system. The bug is _only_ the missing dependency. System W users _must_ get component B installed if they upgrade to version 2 of A. System X users _must_not_ upgrade to version 2 of component A. perhaps you should actually review the bug that blocks this one and the idea of "KEYWORDS" and "DEPENDences" ... when you realize that you cant have a stable package depend on an unstable one, you'll have a EUREKA moment ... until said moment comes about, please stop posting here I'm sorry, but it DOES depend on the unstable mktemp. Not explicity listed in the ebuild, however it *does* *not* work without it. Witness many users having broken systems. If mktemp is unstable, then baselayout-1.12.9 needs to marked unstable too (as of last sync, it is not, amd64 and x86). It really is not good enough to have systems break like this. continue posting all you want, as ive said over and over, the bug that blocks this needs to be resolved first before i make any changes *** Bug 172172 has been marked as a duplicate of this bug. *** fixed in cvs |