Date: Tue, 10 May 2016 14:35:35 +0200 From: Guido Falsi <mad@madpilot.net> To: freebsd-ports@freebsd.org Subject: Re: Poudriere question Message-ID: <b2db2a55-49ae-c4b2-ec10-5ba91b1d5056@madpilot.net> In-Reply-To: <20160510123517.2107653b@gumby.homeunix.com> References: <CAGwOe2Y7HjkK_QxocycmFcKzCUBAVU-87CWqOAzp6ZMUaJMbkA@mail.gmail.com> <3557cbcd-3992-5db5-c5dc-7912508e1956@madpilot.net> <20160510123517.2107653b@gumby.homeunix.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 05/10/16 13:35, RW via freebsd-ports wrote: > On Mon, 9 May 2016 20:15:12 +0200 > Guido Falsi wrote: > >> On 05/09/16 19:52, Fernando Apesteguía wrote: >>> Hi all, >>> >>> Is it safe to use different invocations of poudriere concurrently >>> for different jails but using the same ports collection? >>> >> >> Yes it is, or at least should be. >> >> The ports trees are mounted read only in the jails, the wrkdir is >> defined at a different path. > > What about the distfiles directory? > > Having two "make checksums" running on the same file used to work > fairly well, but not any more because the target now deletes an > incomplete file rather than trying to resume it. > > This wont damage packages, but it can cause two "make checksums" to get > locked in a cycle of deleting each other's files and end with one > getting a failed checksum. Yes it happens, I even have used the same disfiles over NFS with more than one machine/poudriere accessing it. The various instances do overwrite each other and checksums do fail but usually in the end one of them "wins" and the correct file ends up being completed, with other instances reading that one. I agree this happens just by chance and not due to good design. As far as I understand Unix Filesystem semantics each download actually creates a new file, with only the last one to start referencing the actual file visible on the filesystem. So the last one starting to download is the one which will "win" creating the correct file on the FS, then checksumming it and going on. The other files have actuay been deleted and are simply removed from disk as soon as the download ends, if at that point the "winning" one has finished the download, they will checksum that file. There is a chance of the loosing download to end before the winning one ends and overwriting it again, but in my experience with at most 3-4 instances over NFS it usually fixes itself in the long run. IMHO best solution is to make sure you already have distfiles on disk for what you are going to build. -- Guido Falsi <mad@madpilot.net>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?b2db2a55-49ae-c4b2-ec10-5ba91b1d5056>