depot - monorepo for the virus lounge

Age	Commit message (Collapse)	Author	Files	Lines
2013-04-23	Fix --timeout	Eelco Dolstra	1	-38/+25
	I'm not sure if it has ever worked correctly. The line "lastWait = after;" seems to mean that the timer was reset every time a build produced log output. Note that the timeout is now per build, as documented ("the maximum number of seconds that a builder can run").
2013-04-23	Show that --timeout doesn't work if the build produces log output	Eelco Dolstra	2	-3/+1

2013-04-23	nix-build: Respect --timeout	Eelco Dolstra	1	-1/+1

2013-04-23	Nix daemon: respect build timeout from the client	Eelco Dolstra	2	-4/+5

2013-04-23	Fix --fallback with the binary cache substituter	Eelco Dolstra	1	-0/+1
	Reported by Peter Simons.
2013-04-23	Test whether --fallback works if NARS have disappeared from the binary cache	Eelco Dolstra	1	-0/+14

2013-04-23	Test NAR info caching	Eelco Dolstra	2	-5/+15

2013-04-09	Manual: Add a missing step to the build instructions	Eelco Dolstra	1	-3/+4
	Reported by Johan Grande.
2013-04-04	Complain if /homeless-shelter exists	Eelco Dolstra	1	-1/+5

2013-03-25	Fix evaluation of the VM tests	Eelco Dolstra	1	-2/+2

2013-03-25	makeStoreWritable: Ask forgiveness, not permission	Shea Levy	1	-2/+2
	It is surprisingly impossible to check if a mountpoint is a bind mount on Linux, and in my previous commit I forgot to check if /nix/store was even a mountpoint at all. statvfs.f_flag is not populated with MS_BIND (and even if it were, my check was wrong in the previous commit). Luckily, the semantics of mount with MS_REMOUNT \| MS_BIND make both checks unnecessary: if /nix/store is not a mountpoint, then mount will fail with EINVAL, and if /nix/store is not a bind-mount, then it will not be made writable. Thus, if /nix/store is not a mountpoint, we fail immediately (since we don't know how to make it writable), and if /nix/store IS a mountpoint but not a bind-mount, we fail at first write (see below for why we can't check and fail immediately). Note that, due to what is IMO buggy behavior in Linux, calling mount with MS_REMOUNT \| MS_BIND on a non-bind readonly mount makes the mountpoint appear writable in two places: In the sixth (but not the 10th!) column of mountinfo, and in the f_flags member of struct statfs. All other syscalls behave as if the mount point were still readonly (at least for Linux 3.9-rc1, but I don't think this has changed recently or is expected to soon). My preferred semantics would be for MS_REMOUNT \| MS_BIND to fail on a non-bind mount, as it doesn't make sense to remount a non bind-mount as a bind mount.
2013-03-25	makeStoreWritable: Use statvfs instead of /proc/self/mountinfo to find out ↵	Shea Levy	2	-21/+13
	if /nix/store is a read-only bind mount /nix/store could be a read-only bind mount even if it is / in its own filesystem, so checking the 4th field in mountinfo is insufficient. Signed-off-by: Shea Levy <shea@shealevy.com>
2013-03-18	Fix evaluation	Eelco Dolstra	1	-1/+0

2013-03-15	Bump version number	Eelco Dolstra	1	-1/+1

2013-03-15	Remove the "system" jobset input	Eelco Dolstra	1	-14/+18

2013-03-14	Require Bison 2.6	Eelco Dolstra	2	-5/+4

2013-03-14	Fix building against Bison 2.6	Eelco Dolstra	2	-6/+4

2013-03-14	Make sure that thunks are restored properly if an exception occurs	Eelco Dolstra	1	-3/+6
	Fixes Hydra bug #67.
2013-03-08	Prevent config.h from being clobbered	Eelco Dolstra	1	-0/+3

2013-03-08	Revert "Prevent config.h from being clobbered"	Eelco Dolstra	27	-165/+277
	This reverts commit 28bba8c44f484eae38e8a15dcec73cfa999156f6.
2013-03-08	Fix annoying Perl 5.16 warnings	Eelco Dolstra	1	-17/+17
	I.e. Subroutine Nix::Store::isValidPath redefined at /nix/store/clfzsf6gi7qh5i9c0vks1ifjam47rijn-perl-5.16.2/lib/perl5/5.16.2/XSLoader.pm line 92. and so on.
2013-03-07	Prevent config.h from being clobbered	Eelco Dolstra	27	-277/+165

2013-02-28	Handle systems without lutimes() or lchown()	Eelco Dolstra	1	-1/+1

2013-02-28	Update release notes	Eelco Dolstra	2	-1/+11
	Also use a point release version number as suggested by several people.
2013-02-28	Handle symlinks properly	Eelco Dolstra	1	-1/+1
	Now it's really brown paper bag time...
2013-02-28	Bump version number	Eelco Dolstra	1	-1/+1

2013-02-27	Handle hard links to other files in the output	Eelco Dolstra	2	-6/+36

2013-02-27	Refactoring: Split off the non-recursive canonicalisePathMetaData()	Eelco Dolstra	3	-37/+52
	Also, change the file mode before changing the owner. This prevents a slight time window in which a setuid binary would be setuid root.
2013-02-26	Remove outdated file	Eelco Dolstra	1	-33/+0

2013-02-26	Bump version number	Eelco Dolstra	1	-1/+1

2013-02-26	Update release notes	Eelco Dolstra	1	-1/+24

2013-02-26	Security: Don't allow builders to change permissions on files they don't own	Eelco Dolstra	5	-20/+17
	It turns out that in multi-user Nix, a builder may be able to do ln /etc/shadow $out/foo Afterwards, canonicalisePathMetaData() will be applied to $out/foo, causing /etc/shadow's mode to be set to 444 (readable by everybody but writable by nobody). That's obviously Very Bad. Fortunately, this fails in NixOS's default configuration because /nix/store is a bind mount, so "ln" will fail with "Invalid cross-device link". It also fails if hard-link restrictions are enabled, so a workaround is: echo 1 > /proc/sys/fs/protected_hardlinks The solution is to check that all files in $out are owned by the build user. This means that innocuous operations like "ln ${pkgs.foo}/some-file $out/" are now rejected, but that already failed in chroot builds anyway.
2013-02-19	build-remote: Use the --quiet flag	Eelco Dolstra	2	-6/+3
	‘--option verbosity 0’ doesn't actually do anything.
2013-02-19	Add `Settings::nixDaemonSocketFile'.	Ludovic Courtès	5	-10/+14

2013-02-19	Enable chroot support on old glibc versions.	Ludovic Courtès	1	-0/+6

2013-02-08	Document ‘hashString’	Eelco Dolstra	2	-0/+26

2013-02-08	Make "${./path} ..." evaluate to a string, not a path	Eelco Dolstra	3	-5/+7
	Wacky string coercion semantics caused expressions like exec = "${./my-script} params..."; to evaluate to a path (‘/path/my-script params’), because anti-quotations are desuged to string concatenation: exec = ./my-script + " params..."; By constrast, adding a space at the start would yield a string as expected: exec = " ${./my-script} params..."; Now the first example also evaluates to a string.
2013-02-08	Rename "hash" to "hashString" and handle SHA-1	Eelco Dolstra	3	-24/+15

2013-02-08	experimental/hash	Marc Weber	3	-0/+33
	adding primop function calculating hash of a string Signed-off-by: Marc Weber <marco-oweber@gmx.de>
2013-02-05	Nix::Store::derivationFromPath: Return derivation outputs	Eelco Dolstra	1	-9/+12

2013-01-30	Support the colonies	Eelco Dolstra	1	-3/+3

2013-01-24	Improve -I description	Eelco Dolstra	1	-4/+5
	Issue #88.
2013-01-23	Only warn about SQLite being busy once	Eelco Dolstra	1	-1/+5
	No need to get annoying.
2013-01-22	Correctly handle missing logs	Eelco Dolstra	1	-1/+3

2013-01-21	Fix the VM tests	Eelco Dolstra	2	-6/+6

2013-01-21	build-remote.pl: Don't keep build logs on the build slave	Eelco Dolstra	1	-2/+3

2013-01-21	corepkgs/fetchurl: Enable making the downloaded file executable	Shea Levy	1	-3/+4

2013-01-17	Store build logs in /nix/var/log/nix/drvs/<XX>	Eelco Dolstra	2	-30/+37
	...where <XX> is the first two characters of the derivation. Otherwise /nix/var/log/nix/drvs may become so large that we run into all sorts of weird filesystem limits/inefficiences. For instance, ext3/ext4 filesystems will barf with "ext4_dx_add_entry:1551: Directory index full!" once you hit a few million files.
2013-01-07	Bump version number to 1.4	Eelco Dolstra	1	-1/+1

2013-01-05	Delete a left-over trash directory before doing a GC	Eelco Dolstra	1	-1/+4