This is an automated email from the ASF dual-hosted git repository.
alexey pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/kudu.git
The following commit(s) were added to refs/heads/master by this push:
new 24aebba [nvm_cache] set --nvm_cache_allocation_retry_count to 0
24aebba is described below
commit 24aebbaa77f68c813e604eb157b73e0c9bc7433f
Author: Alexey Serbin <[email protected]>
AuthorDate: Thu Oct 17 14:33:49 2019 -0700
[nvm_cache] set --nvm_cache_allocation_retry_count to 0
As it turned out, the logic of retrying NVM-based allocations in
NvmLRUCache::AllocateAndRetry() didn't serve the intended purpose.
The actual memory was deallocated only after all the retries, and
the code effectively was attempting to allocate memory again and
again on nothing-yet-freed-cache.
As a stop-gap solution, this patch sets the default value for the
--nvm_cache_allocation_retry_count flag to 0. The matter will be
properly addressed in a separate changelist.
Change-Id: If33d1a7ffdb7389a02c1b86f2f4b3b9dc362b5d9
Reviewed-on: http://gerrit.cloudera.org:8080/14497
Reviewed-by: Adar Dembo <[email protected]>
Tested-by: Kudu Jenkins
---
src/kudu/util/nvm_cache.cc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/kudu/util/nvm_cache.cc b/src/kudu/util/nvm_cache.cc
index 320048b..66424e3 100644
--- a/src/kudu/util/nvm_cache.cc
+++ b/src/kudu/util/nvm_cache.cc
@@ -60,7 +60,7 @@ DEFINE_string(nvm_cache_path, "/pmem",
"The path at which the NVM cache will try to allocate its
memory. "
"This can be a tmpfs or ramfs for testing purposes.");
-DEFINE_int32(nvm_cache_allocation_retry_count, 10,
+DEFINE_int32(nvm_cache_allocation_retry_count, 0,
"The number of times that the NVM cache will retry attempts to
allocate "
"memory for new entries. In between attempts, a cache entry will
be "
"evicted.");
