Daniel, I notice a gradual creep up in memory used by ganesha when I set up long running load tests. So, I started my tests with mmleak.so preloaded (https://github.com/malahal/mmleak). ( malahal – Compliments on mmleak. Very nice memory debugging tool). After about 15 hours of tests I shrank/processed the information dumped by mmleak. The following is the current counts of the addresses where memory is allocated – without a corresponding free. The first one is reasonable because I have Entries_HWMark = 500000. All locations from Nfs-ganesha V2.7.6 Count Location. Allocation line 500,000 0x52aad0. /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.c:1788 1,700,985 0x539186. /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_hash.h:209 1,700,985 0x53b2b3. /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_helpers.c:392 8,992,812 0x541186. /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_helpers.c:2214 9,015,224 0x538c82. /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_int.h:651 1. Does having 9 million mdcache_dir_entry_t allocated from mdc_readdir_chunk_object (mdcache_helpers.c 2214) and corresponding mdcache_key_dup (mdcache_int.h) seem high? 2. There are 1.7 million entry_export_map (mdcache_helpers.c 392) and cih_hash_key allocations (mdcache_hash.h 209) that as far as I can tell are from mdcache_alloc_entry line 735. Does it seem odd that with 500,000 mdcache entries we have 1.7 million export maps and hash keys? Thoughts? Thanks, Vandana