Why is there an ERANGE (Numerical result out of range) semop error shown?
Issue
We recently made a change in our component liblib_platform_database.so which is responsible for some semop operations in order to guaranty integrity of our SHM segments...
Due to this new change, suddenly we have this error:
17:16:08.158244 semop(304414739, 0x7fffd353d1e0, 1) = -1 ERANGE (Numerical result out of range) <0.000011>
Normally this error happens if the value of the semaphore goes below 0 or exceeds the semvmx value of 32767.
The phenomenon can be reproduced with minimum 2 processes at 3 call/s.
Basically when we really lock a tupple we do -200 in order to have the semaphore to 0.
Environment
- Red Hat Enterprise Linux 5
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.