This should avoid "database is locked" errors when there's a lot of
concurrency, for instance when offloading simultaneously a lot of
builds.
* guix/store/database.scm (call-with-database): Add two 'sqlite-exec'
calls to set 'journal_mode' and 'busy_timeout'.
Previously, if a remote build would fail due to lack of disk space, this
would be considered a permanent failure and thus cached as a build
failure if the local daemon runs with '--cache-failures'.
* guix/scripts/offload.scm (transfer-and-offload): Upon
'nix-protocol-error?' call 'node-free-disk-space' and return 1 instead
of 100 if the result if lower than 10 MiB.
Fixes <https://bugs.gnu.org/33378>.
* guix/scripts/offload.scm (node-free-disk-space): New procedure.
(%minimum-disk-space): New variable.
(choose-build-machine): Call 'node-free-disk-space' and take it into
account in addition to LOAD.
(check-machine-status): Display the free disk space.
* guix/scripts/offload.scm (machine-load): Remove.
(node-load, normalized-load): New procedures.
(choose-build-machine): Call 'open-ssh-session' and 'make-node' from
here; pass the node to 'node-load'.
(check-machine-status): Use 'node-load' instead of 'machine-load'. Call
'disconnect!' on SESSION.
* gnu/system.scm (kernel->boot-label): Get package name and version
using the functions for inferior packages if the kernel is an
inferior package. Return "GNU" if the kernel is not a package.
Signed-off-by: Ludovic Courtès <ludo@gnu.org>
These packages have been missed in
600f1ad8e6, since the `guix refresh
--type=kde` does not honor the sub-directory for the "portingaids".
* gnu/packages/kde-frameworks.scm (kdelibs4support, khtml, kjs,
kjsembed, kmediaplayer, kross): Update to 5.49.0.
* guix/import/cran.scm (fetch-description): Return #F on failure.
(cran->guix-package): Retry from CRAN on failure to fetch description from
bioconductor.
The previous caching strategy, which used STORE as an 'eq?' key, would
no longer work when the functional cache is used because subsequent
store values are not 'eq?'.
* guix/download.scm (built-in-builders*): Rewrite in terms of
'mcached'.
Previously we'd use the (sequence M (map P L)) idiom just because 'mapm'
was slower (not specialized for the given monad). This is no longer the
case since commit dcb95c1fc9.
* guix/gexp.scm (lower-inputs): Use (mapm M P L) instead of (sequence
M (map P L)).
(lower-references, gexp->sexp, imported-files): Likewise.
* guix/profiles.scm (profile-derivation): Likewise.
* guix/scripts/environment.scm (inputs->requisites): Likewise.
* guix/scripts/system.scm (copy-closure): Likewise.
The new release of StumpWM broken the interactive 'gnew' command. It was
fixed upstream [0] and will be part of the next release.
0: https://github.com/stumpwm/stumpwm/pull/538
* gnu/packages/patches/stumpwm-fix-broken-read-one-line.patch: New file.
* gnu/packages/lisp.scm (stumpwm)[source]: Use it.
* gnu/local.mk (dist_patch_DATA): Register it.
Fixes <https://bugs.gnu.org/33751>.
Reported by Marius Bakke <mbakke@fastmail.com>.
* gnu/packages/databases.scm (sqlite-3.26.0): New public variable.
(sqlite)[replacement]: Use it.
(sqlite-with-fts5): Use 'package/inherit'.
(sqlite-with-column-metadata): Likewise.
Modified-by: Mark H Weaver <mhw@netris.org>
The workaround is apparently no longer needed with Linux-libre 4.19.2
in the host.
* gnu/build/vm.scm (load-in-linux-vm): Remove special case for
"qemu-system-i386", thereby re-enabling KVM.
