Re: [trinity-devel] Re: FBTS tdelibs /usr/bin/ld: ../libtdeio.so.14.0.0: file not recognized: file truncated
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512
On 2020/04/24 06:11 AM, deloptes wrote:
On Thu, Apr 23, 2020 at 3:01 AM Michele Calgaro <michele.calgaro@yahoo.it mailto:michele.calgaro@yahoo.it> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512
Hi Michele, yes I am waiting No problem. And yes, there is still the FTBFS and yes I have updated everything to latest and it does not build.
I tried few days ago. Remove everything and run my build script. Failed again on tdelibs.
Give me a note when done, pls.
regards
Hi Emanoil, I have pushed a commit with the updated scripts ytd, but I have not yet a full rebuild. I will do today, then update you again. But do not expect any issue :-)
Which scripts do you mean BTW? your buildscripts? or in the packages?
Hi Emanoil, I was talking about my building scripts. FYI, I did a full rebuild ytd from scratch in bullseye and I had no issues, again :-) Cheers Michele
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Michele Calgaro via trinity-devel wrote:
I was talking about my building scripts. FYI, I did a full rebuild ytd from scratch in bullseye and I had no issues, again :-)
but why am I failing then? I just call debuild. Am I the only one building in debian buster?
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
On Friday 24 of April 2020 13:32:04 deloptes wrote:
Michele Calgaro via trinity-devel wrote:
I was talking about my building scripts. FYI, I did a full rebuild ytd from scratch in bullseye and I had no issues, again :-)
but why am I failing then? I just call debuild. Am I the only one building in debian buster?
I dare say that you are not the only one who builds on Debian Buster - for the stable branch I build for Debian Buster the architectures amd64, i386, ppc64el, arm64, armhf, armel and mips - all successfully.
Cheers
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512
but why am I failing then? I just call debuild. Am I the only one building in debian buster?
I dare say that you are not the only one who builds on Debian Buster - for the stable branch I build for Debian Buster the architectures amd64, i386, ppc64el, arm64, armhf, armel and mips - all successfully.
Cheers
Hi Emanoil, not sure what you are doing wrong, all we can say is that we can build R14.0.8 for all supported debian distros (Slavek) and R14.1.0-dev for bullseye and it works. Slavek and I use different scripts and build method, just as extra info. That points to some sort of problem on your side. - - src code? - - building scripts? - - building environment? - - others?
I would suggest you try building using my scripts as is for a comparison. Then if ok try to work out why yours don't build fine.
Cheers Michele
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, not sure what you are doing wrong, all we can say is that we can build R14.0.8 for all supported debian distros (Slavek) and R14.1.0-dev for bullseye and it works. Slavek and I use different scripts and build method, just as extra info. That points to some sort of problem on your side. - src code?
- building scripts?
- building environment?
- others?
I would suggest you try building using my scripts as is for a comparison. Then if ok try to work out why yours don't build fine.
Very interesting - why its always me :| (joking).
I'll try finding out why it does not build. There must be something missing somewhere. In my script there is nothing special. I borrowed some of the steps from Micheles scripts and created a Makefile with simple set of commands to build. It was a long time ago.
I will have a look, thank you for confirmations that it works on your end.
regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, not sure what you are doing wrong, all we can say is that we can build R14.0.8 for all supported debian distros (Slavek) and R14.1.0-dev for bullseye and it works. Slavek and I use different scripts and build method, just as extra info. That points to some sort of problem on your side. - src code?
- building scripts?
- building environment?
- others?
Hi Michele, all, the problem does not occur when passing -j1 to debuild.
I could not test your scripts, because of pbuilder, which I do not use.
But -j1 seems to be now mandatory for tdelibs and tdebase.
How are you running debuild with -j1 or with -j4?
For me it is interesting to understand if it is i.e. CPU specific - I run it on AMD or similar. I have a note regarding tqca-tls, which also required -j1. Not sure if this still holds, but it seems from time to time it has a problem with -j > 1
thank you and regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
On Monday 27 of April 2020 18:09:49 deloptes wrote:
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, not sure what you are doing wrong, all we can say is that we can build R14.0.8 for all supported debian distros (Slavek) and R14.1.0-dev for bullseye and it works. Slavek and I use different scripts and build method, just as extra info. That points to some sort of problem on your side. - src code?
- building scripts?
- building environment?
- others?
Hi Michele, all, the problem does not occur when passing -j1 to debuild.
I could not test your scripts, because of pbuilder, which I do not use.
But -j1 seems to be now mandatory for tdelibs and tdebase.
How are you running debuild with -j1 or with -j4?
For me it is interesting to understand if it is i.e. CPU specific - I run it on AMD or similar. I have a note regarding tqca-tls, which also required -j1. Not sure if this still holds, but it seems from time to time it has a problem with -j > 1
thank you and regards
I build on several different builders. All are virtual machines in KVM, where some hosts are Intel processor, some AMD processor, one POWER9 processor. The number of cores I have set to 6, 8, 16, 24, 32, 40. So I dare say that I am testing the parallel build quite intensely. And I do not notice problems with building tdelibs.
Note: Now I use sbuild with schroot.
Cheers
Slávek Banko wrote:
I build on several different builders. All are virtual machines in KVM, where some hosts are Intel processor, some AMD processor, one POWER9 processor. The number of cores I have set to 6, 8, 16, 24, 32, 40. So I dare say that I am testing the parallel build quite intensely. And I do not notice problems with building tdelibs.
Note: Now I use sbuild with schroot.
I use it on a simple machine with 4 core and 32GB RAM in a chroot. The chroot is up to date - I mean everything from debian to TDE repository. The only difference I see is that you run it in a VM.
I don't know when but we discussed this -j1 in the past.
Why is this now happening here - has someone ideas?
tdepim just failed too for the same reason. Could it be cache or something?
top - 20:49:45 up 52 days, 22:40, 3 users, load average: 0.05, 0.12, 0.10 Tasks: 314 total, 1 running, 313 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.2 us, 0.4 sy, 0.0 ni, 96.0 id, 3.3 wa, 0.0 hi, 0.1 si, 0.0 st MiB Mem : 32027.2 total, 797.6 free, 604.4 used, 30625.2 buff/cache MiB Swap: 32768.0 total, 32757.5 free, 10.4 used. 30966.7 avail Mem
Perhaps it is worth rebooting and testing again.
regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512
On 2020/04/28 03:50 AM, deloptes wrote:
Slávek Banko wrote:
I build on several different builders. All are virtual machines in KVM, where some hosts are Intel processor, some AMD processor, one POWER9 processor. The number of cores I have set to 6, 8, 16, 24, 32, 40. So I dare say that I am testing the parallel build quite intensely. And I do not notice problems with building tdelibs.
Note: Now I use sbuild with schroot.
I use it on a simple machine with 4 core and 32GB RAM in a chroot. The chroot is up to date - I mean everything from debian to TDE repository. The only difference I see is that you run it in a VM.
I don't know when but we discussed this -j1 in the past.
Why is this now happening here - has someone ideas?
tdepim just failed too for the same reason. Could it be cache or something?
top - 20:49:45 up 52 days, 22:40, 3 users, load average: 0.05, 0.12, 0.10 Tasks: 314 total, 1 running, 313 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.2 us, 0.4 sy, 0.0 ni, 96.0 id, 3.3 wa, 0.0 hi, 0.1 si, 0.0 st MiB Mem : 32027.2 total, 797.6 free, 604.4 used, 30625.2 buff/cache MiB Swap: 32768.0 total, 32757.5 free, 10.4 used. 30966.7 avail Mem
Perhaps it is worth rebooting and testing again.
regards
Hi Emanoil, I also do parallel building, no issues here. So the problem is definitely not there. Have you checked if you are running out of disk space during the build? just speaking out loud here :-$ Cheers Michele
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, I also do parallel building, no issues here. So the problem is definitely= not there. Have you checked if you are running out of disk space during the build? j= ust speaking out loud here :-$ Cheers Michele
No there is 67G free, anyway - I'll reboot next and try.
Thanks for confirming that the error is on my side
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, I also do parallel building, no issues here. So the problem is definitely= not there. Have you checked if you are running out of disk space during the build? j= ust speaking out loud here :-$ Cheers Michele
Hi,
until 56 days ago I was using kernel 4.19.25, but had to upgrade because of a NFS deadlock bug. I upgraded to 4.19.106. Now I upgraded to 4.19.120 and rebooted. Same problem with the parallel building. I wonder if it is the kernel.
I now let it build with -j4 and if fails -j1 debuild -uc -us -b -j4 || debuild -uc -us -b -j1 which seems to work around :/
I will have to dig in further, but I can now build TDE again and try fixing some issues, when time allows.
There is enough disk (60+GB) space BTW on the system.
regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
On Sunday 03 of May 2020 20:18:54 deloptes wrote:
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, I also do parallel building, no issues here. So the problem is definitely= not there. Have you checked if you are running out of disk space during the build? j= ust speaking out loud here :-$ Cheers Michele
Hi,
until 56 days ago I was using kernel 4.19.25, but had to upgrade because of a NFS deadlock bug. I upgraded to 4.19.106. Now I upgraded to 4.19.120 and rebooted. Same problem with the parallel building. I wonder if it is the kernel.
I now let it build with -j4 and if fails -j1 debuild -uc -us -b -j4 || debuild -uc -us -b -j1 which seems to work around :/
I will have to dig in further, but I can now build TDE again and try fixing some issues, when time allows.
There is enough disk (60+GB) space BTW on the system.
regards
Hi Emanoil,
please, you can try this patch?
https://mirror.git.trinitydesktop.org/gitea/TDE/tde-common-cmake/pulls/43
I hope this could solve concurrency issues during the build that could be the source of the FTBFS you are observing. It would be great if it solves the problem you are observing, and I could do a merging of PRs in a common cmake module today.
Cheers
Slávek Banko wrote:
Hi Emanoil,
please, you can try this patch?
https://mirror.git.trinitydesktop.org/gitea/TDE/tde-common-cmake/pulls/43
I hope this could solve concurrency issues during the build that could be the source of the FTBFS you are observing. It would be great if it solves the problem you are observing, and I could do a merging of PRs in a common cmake module today.
Hi Slavek, unfortunately not before the evening. I will try, but have family stuff to take care off.
I'll report if it's working here.
Thanks for looking into it.
regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
Slávek Banko wrote:
On Sunday 03 of May 2020 20:18:54 deloptes wrote:
Michele Calgaro via trinity-devel wrote:
Hi Emanoil, I also do parallel building, no issues here. So the problem is definitely= not there. Have you checked if you are running out of disk space during the build? j= ust speaking out loud here :-$ Cheers Michele
Hi,
until 56 days ago I was using kernel 4.19.25, but had to upgrade because of a NFS deadlock bug. I upgraded to 4.19.106. Now I upgraded to 4.19.120 and rebooted. Same problem with the parallel building. I wonder if it is the kernel.
I now let it build with -j4 and if fails -j1 debuild -uc -us -b -j4 || debuild -uc -us -b -j1 which seems to work around :/
I will have to dig in further, but I can now build TDE again and try fixing some issues, when time allows.
There is enough disk (60+GB) space BTW on the system.
regards
Hi Emanoil,
please, you can try this patch?
https://mirror.git.trinitydesktop.org/gitea/TDE/tde-common-cmake/pulls/43
I hope this could solve concurrency issues during the build that could be the source of the FTBFS you are observing. It would be great if it solves the problem you are observing, and I could do a merging of PRs in a common cmake module today.
Cheers
Hi Slavek, I can confirm it builds now with -j4 without issue.
Thank you! regards
--------------------------------------------------------------------- To unsubscribe, e-mail: trinity-devel-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-devel-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-devel.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
-
deloptes -
Michele Calgaro via trinity-devel -
Slávek Banko