- Package:
- libgrpc++1
- Source:
- grpc
- Description:
- high performance general RPC framework
- Submitter:
- "Michael R. Crusoe"
- Date:
- 2021-06-03 17:21:14 UTC
- Severity:
- important
- Tags:
Hello! As part of creating the new bazel-bootstrap we have run into a bug fixed upstream in 1.30 https://github.com/grpc/grpc/issues/21213 but we woudl like to not have to wait for abseil to pass the NEW queue. Attached is a cherry-picked patch from upstream that fixes the issue. If you could accept it and release 1.26.0-4 with it, then that would be very appreciated! Thanks in advance, - -- System Information: Debian Release: bullseye/sid APT prefers testing APT policy: (650, 'testing'), (600, 'unstable') Architecture: amd64 (x86_64) Foreign Architectures: i386, arm64 Kernel: Linux 5.7.0-1-amd64 (SMP w/8 CPU threads) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE not set Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages libgrpc++1 depends on: ii libc6 2.30-8 ii libgcc-s1 10.1.0-4 pn libgrpc9 <none> ii libprotobuf22 3.11.4-5 pn libprotobuf23 <none> ii libstdc++6 10.1.0-4 ii zlib1g 1:1.2.11.dfsg-2 libgrpc++1 recommends no packages. libgrpc++1 suggests no packages. -----BEGIN PGP SIGNATURE----- iQJGBAEBCgAwFiEEck1gkzcRPHEFUNdHPCZ2P2xn5uIFAl8R6lESHGNydXNvZUBk ZWJpYW4ub3JnAAoJEDwmdj9sZ+biy0EP/jl6/H02JabrQnRCTdfQuTwg6iouu+8r n1zD4695e3OFUBHLFQuOcPsIEjGVqv/N3q6lB1grUGV9oUrjeG5UXpvcIKIx/+V6 tC2UPK7vTAiiw9KcWBobR+M4pEMcCO0XzpH1DRcjvnB3qtnYnsqSIjTfJ2cxVRE/ ssmLd0MZ6YHVCIanEFtWz1dPp4raJJ4UxAm9yOYgPgAm/D1ZDidSZfSgSkJ+xGER 9yBOjRi7hGkQs/6l0FBhUpCwHliWkZTuqyMswaPn3gRREkJdsWLyK7ULn2ZnO8cl KPkxz/PdAlxLfUw6AG27RLEq4ZPEtuNapq3vks3OtbLyB8zZ8ISssrg1aI937Pq5 rcILKzUN/2PiwanxAR3trn0PCMwoV6lJVc0Ipp/85WPzh1WC4clSlfy9lpGm0ymU VX0nRl2Sstej+a1AKYU3mRzLmz2OVGHTheSMdAkxtOcgdCxNAjPjHuEau9J2f1gR It3UGlBd8wjSNcrlvUV7csAOJKWrYRJkG4/3bXCTnltMDtdIGfIs+NpKhzNc6urj CrZV5afIrrDeZkqLp3xklbpAWCDUgdb6KQ5WiRkUyN8Fz3I8sQTHf2OIAprm4ozs X7fx0voBUIaIvyA7o2QXQ2JupzI/tgH/VH1E1SZdy2uCTBOClVS1/KfcQWEbiC1J Xa6qNMrlZM+F =OAEa -----END PGP SIGNATURE-----
Hi, Tried not to bark into abseil packaging anymore, but I think we should move forward. It was already checked by FTP Masters and got rejected. Benjamin fixed that issue and uploaded it. Then again weeks passed. :( Please FTP Masters re-check abseil and accept it to the archives if everything is in order this time. Thanks, Laszlo/GCS
On Sat, Jul 18, 2020 at 11:23 AM László Böszörményi (GCS) <gcs@debian.org> wrote: It has been accepted to the archives, but will fail to build all over in experimental. Reason is simple, Sid has GCC 9.3.0 while experimental has GCC 10.1.0 causing library symbol difference. Benjamin, please upload abseil to Sid when you have time. Thanks, Laszlo/GCS
I just did so. I've successfully built it on all official architectures manually, so I don't expect any build failures, but I'll keep an eye on the builders nonetheless.
Thanks. Most builds came in and failed with symbols difference. Except m68k where the pthreads library can't be found to link with.
I’ve just reuploaded Abseil with an shlibs file instead of a symbols file. The symbols file doesn’t buy us a whole lot anyway, since Abseil is going to break ABI with every release.
It worked, builds are correct this time. Just uploaded gRPC 1.30.2 to unstable now. @Michael: Please test it and report back if it still hangs for you and / or any other patch you need applied.
Hello László, I can confirm the fix. You could add the following as an autopkg test: dependencies: libprotobuf-dev pkg-config protobuf-compiler-grpc protobuf-compiler libgrpc-dev libgrpc++-dev $ cd /examples/cpp/helloworld $ make $ ./greeter_server & ./greeter_client # Server listening on 0.0.0.0:50051 Greeter received: Hello world Thanks! On Fri, Jul 24, 2020 at 6:12 PM László Böszörményi (GCS) <gcs@debian.org> wrote:
Hi! Is there any resolution of this issue? As far as I understand, the version in stable is still 1.26.0-3 which does not include the fix for this problem and I don't know when unstable becomes stable. Is there any intention to update the stable package so that it has the proposed fix? Kind regards, Kirill Bobyrev