Re: Building INDEX fails on 2023Q4
26.10.2023 07:34, Greg Lewis пишет: On 10/25/23 3:52 PM, Vladimir Druzenko wrote: 25.10.2023 21:20, Andrea Venturoli пишет: Hello. I'm experiencing a "make index" failure. Problem seems to be www/tomcat-devel was MFH and requires Java 21+, but 21 is not available in Q4. (20 is the last present, but 18 is the last available from bsd.java.mk). Nothing I can't solve locally (especially since I don't use tomcat-devel :), but I tought I'd let you know. bye & Thanks av. Hi! So we need MFH the java/openjdk21 (2 commits) and Mk/bsd.java.mk (2 commits too). www/tomcat-devel have several CVEs before M13 - can't keep vulnerable version in current quarterly. If you do that, you probably also need to MFH the most recent fix for java/openjdk20 to fix the build on aarch64. Otherwise it won't be able to bootstrap openjdk21 -- Greg Thanks for tip! But I can't do this without approve from java@ and from mentor. Probably it would be faster and easier for someone from java@ to do this, IMHO. -- Best regards, Vladimir Druzenko OpenPGP_0x8006FAABBF942F73.asc Description: OpenPGP public key OpenPGP_signature.asc Description: OpenPGP digital signature
Re: output of RUN_DEPENDS changes unexpectedly
On Tue, Oct 24, 2023, at 11:38 PM, Tatsuki Makino wrote: > Hello. > > I don't know what it is, but as a maintainer but not a committer, I > don't use features related to git branch. > Changes are made directly to files that are checkout-ed, and when > updates are made by pull, etc., stash takes care of everything. > > git stash push --all -- :/ > git pull --autostash ... > git stash pop > git status --show-stash > > Submitted patches are required to have been created with git > format-path, which requires a git commit. > So I do the commit locally, but do a soft reset as soon as the patch is > collected. > > git restore --staged :/ > git add category/portname > git commit > git format-patch -1 HEAD > git reset --soft HEAD^ > > Is there some other way to do it better? :) > Wasn't it like this? :) I don't know. I've never done format-patch or reset when submitting a change. However, I commit changes, so it might be different. Your question might get answers in a new thread. -- Dan Langille d...@langille.org
Re: We need to do something about build times
> On Oct 24, 2023, at 9:12 PM, Robert Clausecker wrote: > > And it seems I'm slowly killing my build SSD like that. After just about > 9 months, it is already at 100 TB of writes just from port builds. > Building with workdirs in memory is no longer an option as that frequently > kills my build server by filling all its RAM with build files until no > processes can be started anymore. Poudriere does not have an effective > mechanism to prevent this (tmpfs limits don't work as the ports in > question require very large workdirs, tend to take very long to build and > tend to be built all at the same time for multiple jails). I will just address a small part of the mail as I am mostly the one maintaining tex. Today I have committed some changes which will reduce the size of the downloads and increase the timing of fetch and install target although it won't be helping too much in the stage and pkg target. In previous I have tried with static plist which often used to timeout more hence I moved into TMPPLIST. Additionally the pkg-plist becomes something like 10MB in size I believe. texlive-texmf is one major requirement for doxygen which is a requirement of many other ports. This is one reason texlive-texmf is a common requirement. If you know of a specific reason this one is rebuilt more frequently let me know I will try to see if the dependency can be avoided to reduce the build frequencies. Kind regards, Moin(bofh@ with all hats off) signature.asc Description: Message signed with OpenPGP
Re: output of RUN_DEPENDS changes unexpectedly
On 24/10/23 15:45, Dan Langille wrote:
Hello,
I noticed an anomaly today.
Watch how the output of make -V RUN_DEPENDS changes for tomcat-devel as we move
through the tree.
This is a fresh checkout of ports
[18:40 empty dan /usr/ports] % cd www/tomcat-devel
[18:41 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
The above is valid output
[18:41 empty dan /usr/ports/www/tomcat-devel] % git checkout 5699269
Note: switching to '5699269'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
git switch -c
Or undo this operation with:
git switch -
Turn off this advice by setting config variable advice.detachedHead to false
HEAD is now at 569926974bfd www/tomcat{85,9,101}: adjust rc.d script to support
Java versions 20+
[18:42 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon /bin/java:
That's not right.
I tried commits before and after 5699269 - both good.
Inquiring minds, FreshPorts not the least, wants to know why the RUN_DEPENDS
failed for that commit.
[18:43 empty dan /usr/ports/www/tomcat-devel] % git checkout d994e83
Previous HEAD position was 569926974bfd www/tomcat{85,9,101}: adjust rc.d
script to support Java versions 20+
HEAD is now at d994e83b45c4 net-mgmt/ocsinventory-{server,ocsreports}: Update
to 2.12.1
[18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
[18:44 empty dan /usr/ports/www/tomcat-devel] % git checkout 2fdb6fb
Previous HEAD position was d994e83b45c4
net-mgmt/ocsinventory-{server,ocsreports}: Update to 2.12.1
HEAD is now at 2fdb6fb45908 www/tomcat{85,9,101}: adjust rc.d script to support
Java versions 20+
[18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
I suspect it's related to the problem reported here:
https://lists.freebsd.org/archives/freebsd-ports/2023-October/004814.html
It seems like when you checkout 5699269 your tree is moved to a point of
quarterly 2023Q4. I see it here as remotes/freebsd/2023Q4~7
Then those java commits are missing and it's not possible to satisfy
dependencies.
--
Renato Botelho
Re: Building INDEX fails on 2023Q4
26.10.2023 13:42 tarihinde Vladimir Druzenko yazdı: 26.10.2023 07:34, Greg Lewis пишет: On 10/25/23 3:52 PM, Vladimir Druzenko wrote: 25.10.2023 21:20, Andrea Venturoli пишет: Hello. I'm experiencing a "make index" failure. Problem seems to be www/tomcat-devel was MFH and requires Java 21+, but 21 is not available in Q4. (20 is the last present, but 18 is the last available from bsd.java.mk). Nothing I can't solve locally (especially since I don't use tomcat-devel :), but I tought I'd let you know. Error is as follows: make_index: /usr/ports/www/tomcat-devel: no entry for /bin/java done. OpenPGP_0x1AD33A94F7FD365D_and_old_rev.asc Description: OpenPGP public key OpenPGP_signature.asc Description: OpenPGP digital signature
Re: output of RUN_DEPENDS changes unexpectedly
On 26/10/23 14:11, Renato Botelho wrote:
On 24/10/23 15:45, Dan Langille wrote:
Hello,
I noticed an anomaly today.
Watch how the output of make -V RUN_DEPENDS changes for tomcat-devel
as we move through the tree.
This is a fresh checkout of ports
[18:40 empty dan /usr/ports] % cd www/tomcat-devel
[18:41 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon
/usr/local/openjdk21/bin/java:java/openjdk21
The above is valid output
[18:41 empty dan /usr/ports/www/tomcat-devel] % git checkout 5699269
Note: switching to '5699269'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
git switch -c
Or undo this operation with:
git switch -
Turn off this advice by setting config variable advice.detachedHead to
false
HEAD is now at 569926974bfd www/tomcat{85,9,101}: adjust rc.d script
to support Java versions 20+
[18:42 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon /bin/java:
That's not right.
I tried commits before and after 5699269 - both good.
Inquiring minds, FreshPorts not the least, wants to know why the
RUN_DEPENDS failed for that commit.
[18:43 empty dan /usr/ports/www/tomcat-devel] % git checkout d994e83
Previous HEAD position was 569926974bfd www/tomcat{85,9,101}: adjust
rc.d script to support Java versions 20+
HEAD is now at d994e83b45c4 net-mgmt/ocsinventory-{server,ocsreports}:
Update to 2.12.1
[18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon
/usr/local/openjdk21/bin/java:java/openjdk21
[18:44 empty dan /usr/ports/www/tomcat-devel] % git checkout 2fdb6fb
Previous HEAD position was d994e83b45c4
net-mgmt/ocsinventory-{server,ocsreports}: Update to 2.12.1
HEAD is now at 2fdb6fb45908 www/tomcat{85,9,101}: adjust rc.d script
to support Java versions 20+
[18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon
/usr/local/openjdk21/bin/java:java/openjdk21
I suspect it's related to the problem reported here:
https://lists.freebsd.org/archives/freebsd-ports/2023-October/004814.html
It seems like when you checkout 5699269 your tree is moved to a point of
quarterly 2023Q4. I see it here as remotes/freebsd/2023Q4~7
Then those java commits are missing and it's not possible to satisfy
dependencies.
I forgot to add on previous email the commit log of 5699269, which has
"Cherry picked from" string on it, what means it's really the commit
from quarterly branch:
commit 569926974bfd179c1806afbe7794aa6d3a73e255 (HEAD)
Author: Vladimir Druzenko
Date: Tue Oct 24 13:28:14 2023 +0300
www/tomcat{85,9,101}: adjust rc.d script to support Java versions 20+
Approved by:arrowd (mentor, implicit)
MFH:2023Q4
(cherry picked from commit 2fdb6fb459089fdb15101c5974b3fa45baef1775)
So if you checkout original commit, 2fdb6fb459, you won't see the problem.
❯ make -V RUN_DEPENDS
jsvc:devel/apache-commons-daemon
/usr/local/openjdk21/bin/java:java/openjdk21
--
Renato Botelho
Re: output of RUN_DEPENDS changes unexpectedly
On Thu, Oct 26, 2023, at 1:11 PM, Renato Botelho wrote:
> On 24/10/23 15:45, Dan Langille wrote:
>> Hello,
>>
>> I noticed an anomaly today.
>>
>> Watch how the output of make -V RUN_DEPENDS changes for tomcat-devel as we
>> move through the tree.
>>
>> This is a fresh checkout of ports
>> [18:40 empty dan /usr/ports] % cd www/tomcat-devel
>> [18:41 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
>> jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
>>
>> The above is valid output
>>
>> [18:41 empty dan /usr/ports/www/tomcat-devel] % git checkout 5699269
>> Note: switching to '5699269'.
>>
>> You are in 'detached HEAD' state. You can look around, make experimental
>> changes and commit them, and you can discard any commits you make in this
>> state without impacting any branches by switching back to a branch.
>>
>> If you want to create a new branch to retain commits you create, you may
>> do so (now or later) by using -c with the switch command. Example:
>>
>>git switch -c
>>
>> Or undo this operation with:
>>
>>git switch -
>>
>> Turn off this advice by setting config variable advice.detachedHead to false
>>
>> HEAD is now at 569926974bfd www/tomcat{85,9,101}: adjust rc.d script to
>> support Java versions 20+
>> [18:42 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
>> jsvc:devel/apache-commons-daemon /bin/java:
>>
>>
>> That's not right.
>>
>> I tried commits before and after 5699269 - both good.
>>
>> Inquiring minds, FreshPorts not the least, wants to know why the RUN_DEPENDS
>> failed for that commit.
>>
>>
>> [18:43 empty dan /usr/ports/www/tomcat-devel] % git checkout d994e83
>> Previous HEAD position was 569926974bfd www/tomcat{85,9,101}: adjust rc.d
>> script to support Java versions 20+
>> HEAD is now at d994e83b45c4 net-mgmt/ocsinventory-{server,ocsreports}:
>> Update to 2.12.1
>> [18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
>> jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
>>
>>
>> [18:44 empty dan /usr/ports/www/tomcat-devel] % git checkout 2fdb6fb
>> Previous HEAD position was d994e83b45c4
>> net-mgmt/ocsinventory-{server,ocsreports}: Update to 2.12.1
>> HEAD is now at 2fdb6fb45908 www/tomcat{85,9,101}: adjust rc.d script to
>> support Java versions 20+
>> [18:44 empty dan /usr/ports/www/tomcat-devel] % make -V RUN_DEPENDS
>> jsvc:devel/apache-commons-daemon /usr/local/openjdk21/bin/java:java/openjdk21
>
> I suspect it's related to the problem reported here:
>
> https://lists.freebsd.org/archives/freebsd-ports/2023-October/004814.html
>
> It seems like when you checkout 5699269 your tree is moved to a point of
> quarterly 2023Q4. I see it here as remotes/freebsd/2023Q4~7
>
> Then those java commits are missing and it's not possible to satisfy
> dependencies.
Good, I understand. As expected, much of the FreshPorts code exists to handle
exceptions such as this. :) I had already coded for this but wanted to follow
up so I understood why it failed.
Thank you.
--
Dan Langille
d...@langille.org
Unmaintained FreeBSD ports which are out of date
Dear port maintainers, The portscout new distfile checker has detected that one or more unmaintained ports appears to be out of date. Please take the opportunity to check each of the ports listed below, and if possible and appropriate, submit/commit an update. Please consider also adopting this port. If any ports have already been updated, you can safely ignore the entry. An e-mail will not be sent again for any of the port/version combinations below. Full details can be found at the following URL: http://portscout.freebsd.org/po...@freebsd.org.html Port| Current version | New version +-+ java/apache-commons-cli | 1.4 | 1.6.0 +-+ If any of the above results are invalid, please check the following page for details on how to improve portscout's detection and selection of distfiles on a per-port basis: http://portscout.freebsd.org/info/portscout-portconfig.txt Reported by:portscout!
Re: We need to do something about build times
Hello. Moin Rahman wrote on 2023/10/27 02:07: > texlive-texmf is one major requirement for doxygen which is a > requirement of many other ports. This is one reason texlive-texmf is a > common requirement. If you know of a specific reason this one is rebuilt > more frequently let me know I will try to see if the dependency can be > avoided to reduce the build frequencies. doxygen is one of the reasons, but so are all ports that require textproc/docbook-utils and print/tex-formats. I integrated a switch in astro/geographiclib that disconnects doxygen, but still works. For audio/libopenshot-audio and multimedia/libopenshot, DOCS option, which is always turned on, would require doxygen. Regards.
