How to use Valgrind to troubleshoot memory leaks in Gluster?
Issue
-
Sometimes, the information provided by an statedump ( please refer to section 19.8 of the Administration Guide regarding the steps on how to capture an statedump) is not sufficient to diagnose the cause of a memory leak.
-
Let's consider the following example, were the structure
nfs3_call_state_t
is allocating 9.3 GB of memory:pool-name=nfs3_call_state_t active-count=143041 sizeof-type=44640 padded-sizeof=65536 size=9374334976 shared-pool=0x7f289a8490e8
-
This structure is being used by many NFS functions in gNFS. How to narrow down which exact function is responsible for this leak?
Environment
Red Hat Gluster Storage 3.X
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.