| Summary: | scripts/bootstrap doesn't work with wrong tempdir | ||
|---|---|---|---|
| Product: | Portage Development | Reporter: | Holger Schurig <holgerschurig> |
| Component: | Unclassified | Assignee: | Sven Vermeulen (RETIRED) <swift> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | carpaski, docs-team, drobbins, everaldo.peixoto |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | x86 | ||
| OS: | All | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
Patch to set "env -i chroot"
Wrong document |
||
|
Description
Holger Schurig
2002-12-23 15:19:46 UTC
Can we just flush the environment? Or make that a note in the install docs? If they have env... env -i chroot /mnt/gentoo /bin/bash Can't just flush the env when inside the chroot. Perhaps an optional script to source? Nick, check in ebuild.sh .. I did add a fix for this some time back, but did not for TEMP. TMP and TMPDIR are covered though. It is in the dyn_compile() and dyn_install() functions. I do not know if we wat to make it more global ... I personally like the idea to change the installation documentation to use "env -i" ... you can never know what envvars are set that would harm the bootstrap or other emerge stuff. Nick, should I fix this on CVS? Is portage CVS up to date ? Do we fix it another way ? *** Bug 16956 has been marked as a duplicate of this bug. *** *** Bug 18701 has been marked as a duplicate of this bug. *** Isn't env available on all systems? If so, I could just change the installation guide to read "env -i chroot /mnt/gentoo /bin/bash" instead of writing a note that people with env should use this and those without use that... Sven, I would expect this to work 100%. K, I'll handle this first thing tomorrow. Created attachment 17500 [details, diff]
Patch to set "env -i chroot"
If anyone can review
I guess env-update takes care of all vars, are you sure you didn't forget it? If azarah, carpaski and drobbins don't say that env-update takes care of this, then I'm rather confident that it doesn't :) Created attachment 17555 [details, diff]
Wrong document
patch looks good. committed |