Frank Middleton wrote:
On 10/ 5/09 01:24 PM, Shawn Walker wrote:
Frank Middleton wrote:
... specifically did
# pkg set-property flush-content-cache-on-success true
There have been no reports of its failure.
Consider it reported! There doesn't seem to be a "pkg get" subcommand
...
to retrieve properties, so it isn't clear that there is a way to prove that
the property is actually set. The above was copy-and-pasted right from
See man pkg(1):
$ pkg property
the command line where it was executed before doing the image-update,
which took two passes to complete successfully (pkg image-update evidently
creates a brand new /rpool/boot/, effectively destroying menu.1st, so it is
impossible to do it in one pass - see bugs 7880 and 6744 which seem unfixed
as of snv111b -> snv124).
Once the image update process starts, pkg(1) operates on a boot
environment clone.
So, if you were expecting the /var/pkg/download in your original boot
environment to be purged, it won't be. However, if you check the new
boot environment's /var/pkg/download, it should be empty ...
Making /var/pkg/download be on a separate dataset is being worked on.
Alternatively, you can set the PKG_CACHEDIR environment variable to a
directory of your choice.
Are these properties and env variables documented anywhere? The man
pages on pkg don't seem to mention these anywhere...
No, these are unofficial. There are RFEs open to document and enhance
control over the download cache.
Cheers,
--
Shawn Walker
_______________________________________________
indiana-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
