FTY remerge from master as of 2024-03-02#2275
FTY remerge from master as of 2024-03-02#2275jimklimov merged 4782 commits intonetworkupstools:FTYfrom
Conversation
|
Side comment: as you remember, that fork is not used anymore nowadays, with the cons of missing dmf support, and the pros of being 2x aligned upstream (using official Debian 12) |
|
Yes, but a lot of treasures lurk there waiting to be upstreamed (and feed back to benefit FTY from the other side). I fear the day when I'd have to review the 50k lines of changes though, and extract that into specific features (DMF, nutconf, vanilla packages, etc.) and more so if trying to keep the 3K items of commit trail :D |
97e475b to
23b0047
Compare
|
@aquette : FWIW, I've hopefully brought Also, finally had a sort of bright idea how to avoid those ugly |
|
Hm, while assigning labels and looking for historic evidence, found the https://github.com/networkupstools/nut/tree/libconf branch among the dozens of loose threads right in this repo. Didn't go that deep into the cellar before :D At least, this gives me a chance to cleanly PR the feature separately of DMF etc. :) |
82d3287 to
6233365
Compare
|
Hm, C++ obsoletions strike again in clang :\ |
6233365 to
6168ce6
Compare
258274d to
7d0b025
Compare
…s first Signed-off-by: Jim Klimov <[email protected]>
…ep-src-docs, avoid sub-makes where we can Signed-off-by: Jim Klimov <[email protected]>
…pared by "git difftool" for comparisons Signed-off-by: Jim Klimov <[email protected]>
Signed-off-by: Jim Klimov <[email protected]>
… and ExecStartPost to monitor that it was applied, from FTY branch Originates from FTY commits: * 062d186 (Jim Klimov 2018-01-16 13:11:08 +0100) * 0e28cc8 (Jim Klimov 2018-01-16 13:37:39 +0100) * e9a6710 (Jim Klimov 2018-01-16 13:41:52 +0100) Signed-off-by: Jim Klimov <[email protected]>
…artLimitInterval for broader compatibility Should help avoid this warning and unachieved functionality on systemd-229 and older: * systemd[1]: [/lib/systemd/system/nut-driver-enumerator-daemon-activator.service:30] Unknown lvalue 'StartLimitIntervalSec' in section 'Service'
…to FTY-remerge-20240123 Signed-off-by: Jim Klimov <[email protected]>
…om-FTY Bump NUT `master` branch from FTY
Signed-off-by: Jim Klimov <[email protected]>
|
Regarding several builds of From warnings analysis summary at https://ci.networkupstools.org/job/nut/job/nut/job/PR-2275/lastBuild/aggregated-analysis/ : Files are indeed differently named and sized:
|
|
So... it seems that the However the "local" one is the package-delivered one: Same as libneon and libnetsnmp, which are packaged and only "local": ...and use system-provided NUT-wise, the warnings seem (parallel build) to appear in these linkages (only for this branch but not for master): The command above does include In both branch builds, libnetsnmp and libneon details were found via pkg-config, so both claim e.g. One notable difference of |
…t clauses, to ease maintenance and navigation Signed-off-by: Jim Klimov <[email protected]>
…p rules Signed-off-by: Jim Klimov <[email protected]>
|
Tested, libxml2 was not the culprit (makes sense, little use for crypto there). Found that DMF branch also includes this sort of difference: and the link commands in build logs reflect that ( (they were built into The openssl-1.1.1 libraries are honestly found via |
… "pkg add openssl" in fact [networkupstools#2275] Signed-off-by: Jim Klimov <[email protected]>
… "pkg add openssl" in fact [networkupstools#2275] Signed-off-by: Jim Klimov <[email protected]>
…fy the unit description (it does not strictly restart the daemon, can reload if running too) Signed-off-by: Jim Klimov <[email protected]>
…om-FTY-systemd Bump NUT `master` branch from FTY: systemd integration
Signed-off-by: Jim Klimov <[email protected]>
…s to running the ./configure script Signed-off-by: Jim Klimov <[email protected]>
Signed-off-by: Jim Klimov <[email protected]>
…lkp_default(), snmp_info_sentinel and info_lkp_sentinel macros to simplify the codebase [networkupstools#2275] Signed-off-by: Jim Klimov <[email protected]>
Bump the copy of 42ity fork contents to minimize difference with the master (as they got to differ considerably over the past 2 years of main trunk's evolution). Hopefully this would ease eventual import of features from the fork and facilitate comparisons of intentional differences.
Also letting NUT CI chew it in many more scenarios than I can run locally :)
After all, it is very possible that difftool/mergetool duplicated or ate away some lines (fixed some cases like that which did break local builds soon after initial conflict resolution).