Activity for Thien Minh Huynh
-
Thien Minh Huynh modified ticket #3360
imm: multiple value attribute is out of order when load xml file
-
imm: fix multiple value out of order when import from xml file [#3360]
-
imm: multiple value attribute is out of order when load xml file
-
amf: Update amf clc cli with dynamic buffer [#3358]
-
ntf: ntfa coredump when use saNtfFinalize api
-
commit 7edd13a68302344a5e552e9e28c04c7a882fe684 (HEAD -> develop, origin/develop, ticket-3357) Author: thien.m.huynh thien.m.huynh@endava.com Date: Wed Aug 14 14:10:45 2024 +0700 ntf: test case to test manage ntfa resource [#3357] commit 862493e85ee889bbb0d2369e357d9360214bc39f Author: thien.m.huynh thien.m.huynh@endava.com Date: Wed Aug 14 14:04:18 2024 +0700 ntf: fix thread unsafe [#3357] In case finalize get the head pointer before mutex lock. That leads to the iterator operation corrupt. This...
-
ntf: ntfa coredump when use saNtfFinalize api
-
ntf: test case to test manage ntfa resource [#3357]
-
ntf: fix thread unsafe [#3357]
-
amf: amfd coredump with buffer overflow
-
ntf: ntfa coredump when use saNtfFinalize api
-
imm: introduce new option for immcfg that allow overwrite when load from xml file
-
pyosaf: python3 SaStringT compare with str alway fail
-
commit 321e1d9eeac01d5259c73c88cea56e34462ebe8e (HEAD -> develop, origin/develop, ticket-3351) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Tue Apr 23 10:16:38 2024 +0700 pyosaf: Allow SaStringT compare with str by default [#3351] An earlier patch supporting Python 3 introduced the SaStringT class. With str type, default SaStringT cannot be compared. This commit is to allow compare with str class if without casting
-
pyosaf: Allow SaStringT compare with str by default [#3351]
-
pyosaf: python3 SaStringT compare with str alway fail
-
pyosaf: Allow SaStringT compare with str by default [#3351]
-
pyosaf: python3 SaStringT compare with str alway fail
-
pyosaf: python3 SaStringT compare with str alway fail
-
amf: component can not recovery while imm unavailable
-
commit f8fd609eb95fa347b0f1f94cb73bbd94fac61229 (HEAD -> develop, origin/develop, ticket-3344) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Fri Dec 29 10:41:06 2023 +0700 amf: retry if imm unavailable during recovery [#3344] If a component has changed, AMF will obtain configuration from IMM when recovering it. If IMM is unavailable and SC absence is enabled, AMF will immediately fail. Retrying the cleanup event is the solution. This is to prevent watchdog kill AMFND.
-
amf: retry if imm unavailable during recovery [#3344]
-
amf: amfd crash during remove amf model
-
commit 3463fa7734dab0b00365e6665eee4f896ed6268e (HEAD -> develop, origin/develop, ticket-3345) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Fri Dec 29 13:14:58 2023 +0700 amf: check the existing SG before creating a new one [#3345] When IMM is restarted, AMF will be re-initialized. AMF add duplicate AMF SG entities to the local database. It causes crashing while trying to delete an App object. Solution is to check for existing SG before create a new one.
-
amf: check the existing SG before creating a new one [#3345]
-
amf: component can not recovery while imm unavailable
-
amf: retry if imm unavailable during recovery [#3344]
-
amf: amfd crash during remove amf model
-
amf: check the existing sg before create new one [#3345]
-
amf: amfd crash during remove amf model
-
amf: component can not recovery while imm unavailable
-
amf: component can not recovery while imm unavailable
-
amf: component can not recovery while imm unavailable
-
log: memleak detected by valgrind
-
mbc: fix memleak detected by valgrind [#3341]
-
log: memleak detected by valgrind
-
pyosaf: correct type before cast to a ctypes value pointer [#3338]
-
mds: fix coredump when run mdstest using mds q_ownership [#3337]
-
evt: fix pointer used after free [#3332]
-
rde: fix pointer used after free [#3332]
-
rde: fix errors reported by Valgrind [#3333]
-
log: osaflogd coredump after receiving a peer checkpoint
-
commit 47dee89190f1be5714010b5e546dc6a3168a74f0 (HEAD -> develop, origin/develop, ticket-3334) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Thu Apr 6 18:07:26 2023 +0700 log: check the queue capacity before retrieving an item [#3334] Osaflogd occasionally fails to keep synchronization in a pair of nodes. When received a pop request from active, standby will be coredump if retrieving a piece from an empty queue. This is done to avoid coredump by checking the queue capacity before retrieving...
-
log: osaflogd coredump after receiving a peer checkpoint
-
log: check the queue capacity before retrieving an item [#3334]
-
log: osaflogd coredump after receiving a peer checkpoint
-
log: osaflogd coredump after receiving a peer checkpoint
-
clm: Add an operation to excute a script [#2696]
-
clm: Update the clm-adm command-line tool to support new admin-ops
-
amf: Fix errors reported by valgrind (cont) [#3331]
-
imm: Make NFS state consistent among nodes [#3330]
-
dtm: flush the logtrace asap when the logtrace owner is terminated
-
commit 9c34b746228c65f96bb894082c910ac59cc951ef (HEAD -> develop, origin/develop, ticket-3106) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Fri Oct 25 15:18:06 2019 +0700 dtm: flush logtrace asap when logtrace owner is terminated [#3106] To avoid the risk of losing traces, logtrace server should flush traces as soon as possible when logtrace owner is terminated. This patch also refactors LogServer::Run().
-
dtm: flush logtrace asap when logtrace owner is terminated [#3106]
-
ntf: Syscall param socketcall.sendto(msg) points to uninitialised byte(s)
-
commit ae082e80cb8c4ea958f51bf3e4b6a61e9ec4702b (HEAD -> develop, origin/develop, ticket-3326) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Wed Dec 7 16:31:39 2022 +0700 log: fix access uninitialised value [#3326] Struct SaVersionT has only 24bit but using encode 32bit. Solution is encode each 8bit releaseCode, majorCode, minorCode by using ncs_edp_saversiont. And encode one more 8bit to keep backward compatible. commit e6f3d67fde081fcd60c5759436a7b90251952ae3 Author: thien.m.huynh thien.m.huynh@dektech.com.au...
-
imm: fix access uninitialised value [#3326]
-
log: fix access uninitialised value [#3326]
-
mbc: fix access uninitialised value [#3326]
-
ntf: fix access uninitialised value [#3326]
-
clm: fix access uninitialised value [#3326]
-
log: fix access uninitialised value [#3326]
-
mbc: fix access uninitialised value [#3326]
-
ntf: fix access uninitialised value [#3326]
-
imm: fix access uninitialised value [#3326]
-
clm: fix access uninitialised value [#3326]
-
ntf: Syscall param socketcall.sendto(msg) points to uninitialised byte(s)
-
ntf: Syscall param socketcall.sendto(msg) points to uninitialised byte(s)
-
log: fix access uninitialised value [#3326]
-
imm: fix access uninitialised value [#3326]
-
mbc: fix access uninitialised value [#3326]
-
ntf: fix access uninitialised value [#3326]
-
clm: fix access uninitialised value [#3326]
-
log: Invalid read is detected
-
commit 2021d0c035d2989267ee8d0d8d826ddf53fc3201 (HEAD -> develop, origin/develop, ticket-3325) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Fri Nov 25 13:12:42 2022 +0700 log: fix invalid read [#3325] memcpy copies invalid memory that has not yet been initialized. So this fix is the correct size when using memcpy.
-
log: fix invalid read [#3325]
-
ntf: Syscall param socketcall.sendto(msg) points to uninitialised byte(s)
-
log: Invalid read is detected
-
dtm: flush the logtrace asap when the logtrace owner is terminated
-
dtm: flush logtrace asap when logtrace owner is terminated [#3106]
-
immnd: unexpected reboot node after merge network back
-
Hi Mohan, On behalf of Huu, I would like to share some information on the issue. The issue occurs when the network is split into partitions: partition1 = [SC-1, SC-2, SC-3, SC-4, SC-5] partition2 = [SC-6, SC-7, SC-8, SC-9, SC-10] I am not able to reproduce the issue. And the issue has not been seen again in a long time. I will close ticket as not reproduce. Best Regards, Thien
-
imm: multiple value attribute is out of order after cluster reboot
-
imm: multiple value attribute is out of order after cluster reboot
-
commit 3beb5ab488600d4e4978084b8ea801a6de60ab15 (HEAD -> develop, origin/develop, ticket-3321) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Tue Sep 20 17:13:00 2022 +0700 imm: fix multiple value attribute out of order [#3321] Create an object that has multiple value attributes. After that, IMMND send the create callback to IMMPBED. But the value is sent in the same order as when it was created. IMMND will load data from immpbed after the cluster reboots. That leads to multiple values...
-
imm: multiple value attribute is out of order after cluster reboot
-
imm: fix multiple value attribute out of order [#3321]
-
imm: multiple value attribute is out of order after cluster reboot
-
log: delete the empty log file when rotate log file
-
commit 0b96adb085209e43bcb643b4c998f43c283c30e4 (HEAD -> develop, origin/develop, ticket-3320) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Thu Aug 4 08:59:34 2022 +0700 log: delete the empty log file when rotate log file [#3320] When performing an admin op command with id = 2 (SA_LOG_ADMIN_ROTATE_FILE) without parameters and there are no log record in the stream. There will be a empty log file created. By checking the size of the stream, this ticket avoids creating an empty log file....
-
log: delete the empty log file when rotate log file [#3320]
-
log: delete the empty log file when rotate log file
-
log: delete the empty log file when rotate log file [#3320]
-
log: delete the empty log file when rotate log file
-
commit 12553142ea53b5ea1b608cdac64973a50dd231ee (HEAD -> develop, origin/develop, ticket-3316) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Tue Jun 7 13:08:32 2022 +0700 base: Correctly use the MAX_GRP_MEM_BUF_SIZE [#3316] Reset errno to prevent ERANGE unexpected. Align realloc memory with MAX_GRP_MEM_BUF_SIZE. Improve logging for unexpected configure.
-
base: Correctly use the MAX_GRP_MEM_BUF_SIZE [#3316]
-
base: improve the logging and reset errno [#3316]
-
dtm: fix osaflog tool to work with short argument [#3295]
-
imm: immnd coredump during authenticate non-local user
-
commit e5a19e7770b1ad3a99af88252dddec6b7f875b54 (HEAD -> develop, origin/develop, ticket-3305) Author: thien.m.huynh thien.m.huynh@dektech.com.au Date: Tue Feb 8 13:38:42 2022 +0700 base: fix immnd crash when authenticate non-local user [#3305] Correct length of buffer and init memory when allocate memory.
1 >