Thread: semctl(): Invalid argument
What would cause the following? # psql -U pgsql -l psql: FATAL: semctl(327686, 3, SETVAL, 0) failed: Invalid argument The server is FreeBSD 6.1-PRERELEASE ... there are two jails running on it ... both jails are running PostgreSQL 7.4 ... I have the kernel built with enough semaphores: options SYSVSHM options SHMMAXPGS=199608 options SHMMAX=(SHMMAXPGS*PAGE_SIZE+1) options SYSVSEM options SEMMNI=4096 options SEMMNS=8192 options SYSVMSG # SYSV-style message queues And am definitely not using that many: (from ipcs -a): ID KEY MODE OWNER GROUP CREATOR CGROUP NSEMS OTIME CTIME s 393216 5432001 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393217 5432002 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393218 5432003 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393219 5432004 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393220 5432005 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393221 5432006 --rw------- 70 70 70 70 17 13:46:07 13:46:07 s 393222 5432007 --rw------- 70 70 70 70 17 13:46:07 13:46:07 So figure there must be something else that I should be looking at ... Thanks ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
It looks like moving to FreeBSD 6.x for some of our servers is biting me in the butt ... they "broke" something with semaphores that used to work with 4.x, so that you can't run two postmasters on port 5432 within two seperate jails ... :( On Sun, 2 Apr 2006, Marc G. Fournier wrote: > > What would cause the following? > > # psql -U pgsql -l > psql: FATAL: semctl(327686, 3, SETVAL, 0) failed: Invalid argument > > The server is FreeBSD 6.1-PRERELEASE ... there are two jails running on it > ... both jails are running PostgreSQL 7.4 ... I have the kernel built with > enough semaphores: > > options SYSVSHM > options SHMMAXPGS=199608 > options SHMMAX=(SHMMAXPGS*PAGE_SIZE+1) > > options SYSVSEM > options SEMMNI=4096 > options SEMMNS=8192 > > options SYSVMSG # SYSV-style message queues > > And am definitely not using that many: > > (from ipcs -a): > ID KEY MODE OWNER GROUP CREATOR CGROUP > NSEMS OTIME CTIME s 393216 5432001 --rw------- 70 > 70 70 70 17 13:46:07 13:46:07 > s 393217 5432002 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > s 393218 5432003 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > s 393219 5432004 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > s 393220 5432005 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > s 393221 5432006 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > s 393222 5432007 --rw------- 70 70 70 70 > 17 13:46:07 13:46:07 > > So figure there must be something else that I should be looking at ... > > Thanks ... > > ---- > Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) > Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664 > > ---------------------------(end of broadcast)--------------------------- > TIP 6: explain analyze is your friend > ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
FWIW, i've been running two postmasters on the same port on a FreeBSD 6.0-RELENG server, one in the host and one in a jail (this is on a devel server). I don't like running postgresql in a jail anyway, because it involves enabling sysv IPC inside jails, which sort of defeats the entire purpose of having a jail.. Gr, Koen Marc G. Fournier wrote: > > It looks like moving to FreeBSD 6.x for some of our servers is biting me > in the butt ... they "broke" something with semaphores that used to work > with 4.x, so that you can't run two postmasters on port 5432 within two > seperate jails ... :( > > On Sun, 2 Apr 2006, Marc G. Fournier wrote: > >> >> What would cause the following? >> >> # psql -U pgsql -l >> psql: FATAL: semctl(327686, 3, SETVAL, 0) failed: Invalid argument >> >> The server is FreeBSD 6.1-PRERELEASE ... there are two jails running >> on it ... both jails are running PostgreSQL 7.4 ... I have the kernel >> built with enough semaphores: >> >> options SYSVSHM >> options SHMMAXPGS=199608 >> options SHMMAX=(SHMMAXPGS*PAGE_SIZE+1) >> >> options SYSVSEM >> options SEMMNI=4096 >> options SEMMNS=8192 >> >> options SYSVMSG # SYSV-style message queues >> >> And am definitely not using that many: >> >> (from ipcs -a): >> ID KEY MODE OWNER GROUP CREATOR >> CGROUP NSEMS OTIME CTIME s 393216 5432001 >> --rw------- 70 70 70 70 17 13:46:07 13:46:07 >> s 393217 5432002 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> s 393218 5432003 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> s 393219 5432004 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> s 393220 5432005 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> s 393221 5432006 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> s 393222 5432007 --rw------- 70 70 >> 70 70 17 13:46:07 13:46:07 >> >> So figure there must be something else that I should be looking at ... >> >> Thanks ... >> >> ---- >> Marc G. Fournier Hub.Org Networking Services >> (http://www.hub.org) >> Email: scrappy@hub.org Yahoo!: yscrappy ICQ: >> 7615664 >> >> ---------------------------(end of broadcast)--------------------------- >> TIP 6: explain analyze is your friend >> > > ---- > Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) > Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664 > > ---------------------------(end of broadcast)--------------------------- > TIP 6: explain analyze is your friend > > -- K.F.J. Martens, Sonologic, http://www.sonologic.nl/ Networking, hosting, embedded systems, unix, artificial intelligence. Public PGP key: http://www.metro.cx/pubkey-gmc.asc Wondering about the funny attachment your mail program can't read? Visit http://www.openpgp.org/