test: create the packpack suse leap 15.[0..2] images and testing with it #4562

avtikhon · 2019-10-11T07:43:43Z

OS version:
suse 15.[0-2]

Blocked by:

Added makefile to build the SuSE based packages. Part of #4562

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Closes #4562

avtikhon · 2020-04-10T06:01:34Z

Waiting sub-task at:
packpack/packpack#121

ligurio · 2020-04-23T07:05:37Z

@avtikhon please add web link to patch and move to review back

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Closes #4562

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites and tests in luajit repository. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Closes #4562 t

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites and tests in luajit repository. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Close #4562

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites and tests in luajit repository. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs with testings. Close #4562

Added makefile to build the SuSE based packages. Implemented build with testing for opensuse images: SuSE 15.0, 15.1, 15.2 Tarantool spec file changed with: Added checks of %{sle_version} according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Found that opensuse adding linker flag like '--no-undefined' which produces the fails on building test modules at app-tap/app/box-tap suites and tests in luajit repository. Decided to block this flag. Running tests by make test, because opensuse uses out-of-source build by default. Set complete testing except replication suite. Test box-tap/cfg fails and temporary blocked according to the issue created for it #4594. Added rpm/prebuild-opensuse-leap.sh script to fix the issue with broken repositories in zypper cache. It refreshes zypper cache. Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs with testings. Close #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behaviour was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno whould be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself than the check was wrong. The previous logic of the check was to catch the error ENOENT from inside the xdir_scan() function to handle the situation when vinyl_dir was not exist. In this way errno should be reseted before xdir_scan() call to give the ability for xdir_scan() to use return value without errno set and correctly handle errno from inside the xdir_scan(). After discussions found that there was alternative better solution to fix it. To be sure in behaviour of the changing errno decided to use variant with the flag in xdir_scan() if the directory should exist. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behaviour was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno whould be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself than the check was wrong. The previous logic of the check was to catch the error ENOENT from inside the xdir_scan() function to handle the situation when vinyl_dir was not exist. In this way errno should be reseted before xdir_scan() call to give the ability for xdir_scan() to use return value without errno set and correctly handle errno from inside the xdir_scan(). After discussions found that there was alternative better solution to fix it. As mentioned above xdir_scan() function is not system call and can be changed inside it in any possible way. So check outside of this function on errno could be broken, because of the xdir_scan() changes. To avoid of it we must avoid of errno checks outside of the function. Better solution was to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself than the check was wrong. The previous logic of the check was to catch the error ENOENT from inside the xdir_scan() function to handle the situation when vinyl_dir was not exist. In this way errno should be reset before xdir_scan() call to give the ability for xdir_scan() to use return value without errno set and correctly handle errno from inside the xdir_scan(). After discussions found that there was alternative better solution to fix it. As mentioned above xdir_scan() function is not system call and can be changed inside it in any possible way. So check outside of this function on errno could be broken, because of the xdir_scan() changes. To avoid of it we must avoid of errno checks outside of the function. Better solution was to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself than the check was wrong. The previous logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. In this way errno should be reset before xdir_scan() call to give the ability for xdir_scan() to use return value without errno set and correctly handle errno which could be set in the xdir_scan(). After discussions found that there was alternative better solution to fix it. As mentioned above xdir_scan() function is not system call and can be changed inside it in any possible way. So check outside of this function on errno could be broken, because of the xdir_scan() changes. To avoid of it we must avoid of errno checks outside of the function. Better solution was to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself than the check was wrong. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it errno had to be reset before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan() and success system calls in xdir_scan() hadn't clean it up. As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid of it we must avoid of errno checks outside of the function. Better solution was to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself it produced the wrong check. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it possible fix could be reset errno before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan(). As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid of it we must avoid of errno checks outside of the function. Better solution is to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself it produced the wrong check. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it possible fix could be reset errno before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan(). As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid that we must not check errno after call of the function. Better solution is to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself it produced the wrong check. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it possible fix could be reset errno before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan(). As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid that we must not check errno after call of the function. Better solution is to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org> (cherry picked from commit 77367c0)

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562 (cherry picked from commit edf622b)

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself it produced the wrong check. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it possible fix could be reset errno before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan(). As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid that we must not check errno after call of the function. Better solution is to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org> (cherry picked from commit 43c776a)

During implementation of openSUSE build with testing got failed test box-tap/cfg.test.lua. Found that when memtx_dir didn't exist and vinyl_dir existed and also errno was set to ENOENT, box configuration succeeded, but it shouldn't. Reason of this wrong behavior was that not all of the failure paths in xdir_scan() set errno, but the caller assumed it. Debugging the issue found that after xdir_scan() there was incorrect check for errno when it returned negative values. xdir_scan() is not system call and negative return value from it doesn't mean that errno would be set too. Found that in situations when errno was left from previous commands before xdir_scan() and xdir_scan() returned negative value by itself it produced the wrong check. The previous failed logic of the check was to catch the error ENOENT which set in the xdir_scan() function to handle the situation when vinyl_dir was not exist. It failed, because checking ENOENT outside the xdir_scan() function, we had to be sure that ENOENT had come from xdir_scan() function call indeed and not from any other functions before. To be sure in it possible fix could be reset errno before xdir_scan() call, because errno could be passed from any other function before call to xdir_scan(). As mentioned above xdir_scan() function is not system call and can be changed in any possible way and it can return any result value without need to setup errno. So check outside of this function on errno could be broken. To avoid that we must not check errno after call of the function. Better solution is to use the flag in xdir_scan(), to check if the directory should exist. So errno check was removed and instead of it the check for vinyl_dir existence using flag added. Closes #4594 Needed for #4562 Co-authored-by: Alexander Turenko <alexander.turenko@tarantool.org>

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562

Implemented openSUSE packages build with testing for images: opensuse-leap:15.[0-2] Added %{sle_version} checks in Tarantool spec file according to https://en.opensuse.org/openSUSE:Packaging_for_Leap#RPM_Distro_Version_Macros Added opensuse-leap of 15.1 and 15.2 versions to Gitlab-CI packages building/deploing jobs. Closes #4562 (cherry picked from commit d07e5f9)

avtikhon added feature A new functionality qa Issues related to tests or testing subsystem labels Oct 11, 2019

avtikhon self-assigned this Oct 11, 2019

avtikhon added a commit that referenced this issue Oct 21, 2019

build: add zypper makefile for SuSE

63545a4

Added makefile to build the SuSE based packages. Part of #4562

kyukhin added this to the 2.3.1 milestone Nov 8, 2019

kyukhin modified the milestones: 2.3.1, 2.3.2 Dec 30, 2019

avtikhon added this to ON REVIEW in Quality Assurance Apr 7, 2020

avtikhon mentioned this issue Apr 10, 2020

test: box-tap/cfg test fails on SuSE 15.x images #4594

Closed

kyukhin modified the milestones: 2.3.2, 2.3.3 Apr 21, 2020

ligurio moved this from ON REVIEW to DOING in Quality Assurance Apr 23, 2020

avtikhon moved this from DOING to ON REVIEW in Quality Assurance Apr 23, 2020

kyukhin modified the milestones: 2.3.3, QA Apr 23, 2020

avtikhon moved this from ON REVIEW to ON REVIEW WITH 1 LGTM in Quality Assurance Aug 21, 2020

avtikhon moved this from ON REVIEW WITH 1 LGTM to 2 LGTMs ready to push in Quality Assurance Aug 31, 2020

avtikhon moved this from 2 LGTMs ready to push to ON REVIEW WITH 1 LGTM in Quality Assurance Aug 31, 2020

kyukhin closed this as completed in d07e5f9 Aug 31, 2020

avtikhon moved this from ON REVIEW WITH 1 LGTM to DONE in Quality Assurance Aug 31, 2020

avtikhon removed this from DONE in Quality Assurance Sep 15, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

test: create the packpack suse leap 15.[0..2] images and testing with it #4562

test: create the packpack suse leap 15.[0..2] images and testing with it #4562

avtikhon commented Oct 11, 2019 •

edited

avtikhon commented Apr 10, 2020 •

edited

ligurio commented Apr 23, 2020

test: create the packpack suse leap 15.[0..2] images and testing with it #4562

test: create the packpack suse leap 15.[0..2] images and testing with it #4562

Comments

avtikhon commented Oct 11, 2019 • edited

avtikhon commented Apr 10, 2020 • edited

ligurio commented Apr 23, 2020

avtikhon commented Oct 11, 2019 •

edited

avtikhon commented Apr 10, 2020 •

edited