Couchbase Server

erlang FULL_SWEEP setting needs override to 512 instead of erlang's default

Details

  • Type: Bug Bug
  • Status: Resolved Resolved
  • Priority: Blocker Blocker
  • Resolution: Fixed
  • Affects Version/s: 2.0-beta
  • Fix Version/s: 2.0-beta-2
  • Component/s: ns_server
  • Security Level: Public
  • Labels:
    None

Description

Looks like 512 is the better setting than the default...

  https://docs.google.com/spreadsheet/ccc?key=0AgLUessE73UXdGd2NmhLOThCRUc0ekFkZ0FDeGdLRXc#gid=0

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Steve Yen added a comment -
Changing title to be clearer.

Also, I spoke with Damien and he's good with 512, too.
Show
Steve Yen added a comment - Changing title to be clearer. Also, I spoke with Damien and he's good with 512, too.
Hide
Permalink
Pavel Paulau added a comment -
You don't wait for system test results? You don't need extra perf runs for that?
Show
Pavel Paulau added a comment - You don't wait for system test results? You don't need extra perf runs for that?
Hide
Permalink
Aleksey Kondratenko added a comment -
We can always revert later. But I'd wait at least a bit.
Show
Aleksey Kondratenko added a comment - We can always revert later. But I'd wait at least a bit.
Hide
Permalink
Farshid Ghods added a comment -
Tommie,

can you update the ticket once you have initial results from pine and plum cluster where we are running the cluster with this settings ?
Show
Farshid Ghods added a comment - Tommie, can you update the ticket once you have initial results from pine and plum cluster where we are running the cluster with this settings ?
Hide
Permalink
Tommie McAfee added a comment -
Current status of verification on key-value and query clusters:
 
Overall looks good on 23 node with key-value workload with SWEEP=512:
 * load 20 Million items, size 356bytes
 * access phase: --get 70% --create 10% --update 15% --delete 5% --ops 40000
 * rebalance out 3 nodes completed
 * load until 44% active resident
 * access phase with cache_miss ratio between 1-3%
 * swap rebalance in 3 nodes, out 3 nodes completed

Timeouts on 4 node cluster with views and SWEEP=10000:
  * load 13 million items into 2 different buckets
  * run 300/queries-per-sec against each bucket
  * access phase at 70% active resident
  * rebalance out 1 node
        - Failed with etimeout
        - usually able to retry and will succeed

On 4 node cluster with views and SWEEP=512:
   * verification pending...
Show
Tommie McAfee added a comment - Current status of verification on key-value and query clusters:   Overall looks good on 23 node with key-value workload with SWEEP=512:  * load 20 Million items, size 356bytes  * access phase: --get 70% --create 10% --update 15% --delete 5% --ops 40000  * rebalance out 3 nodes completed  * load until 44% active resident  * access phase with cache_miss ratio between 1-3%  * swap rebalance in 3 nodes, out 3 nodes completed Timeouts on 4 node cluster with views and SWEEP=10000:   * load 13 million items into 2 different buckets   * run 300/queries-per-sec against each bucket   * access phase at 70% active resident   * rebalance out 1 node         - Failed with etimeout         - usually able to retry and will succeed On 4 node cluster with views and SWEEP=512:    * verification pending...
Hide
Permalink
Steve Yen added a comment -
assigning back to alk to get the 512 setting changed.
Show
Steve Yen added a comment - assigning back to alk to get the 512 setting changed.
Hide
Permalink
Tommie McAfee added a comment -

Recent system test verification failed due to another issue MB-6490….Instead of the timeouts we saw at higher gc levels.
Show
Tommie McAfee added a comment - Recent system test verification failed due to another issue MB-6490 ….Instead of the timeouts we saw at higher gc levels.
Hide
Permalink
Steve Yen added a comment -
changing priority to blocker (from critical)
Show
Steve Yen added a comment - changing priority to blocker (from critical)
Hide
Permalink
Aleksey Kondratenko added a comment -
Change is in gerrit and will be merged soon
Show
Aleksey Kondratenko added a comment - Change is in gerrit and will be merged soon
Hide
Permalink
Steve Yen added a comment -
Aliaksey A. had a good point on the fix, where we need a corresponding fix for the windows?
Show
Steve Yen added a comment - Aliaksey A. had a good point on the fix, where we need a corresponding fix for the windows?
Hide
Permalink
Karen Zeller added a comment -
added to RN: By default we provide garbage collection more frequently than the
normal default for Erlang. This keeps memory usage by the Erlang
virtual machine lower, and enables better performance.
Show
Karen Zeller added a comment - added to RN: By default we provide garbage collection more frequently than the normal default for Erlang. This keeps memory usage by the Erlang virtual machine lower, and enables better performance.

People

  • Assignee:
    Aleksey Kondratenko
    Reporter:
    Steve Yen
Vote (0)
Watch (0)

Dates

  • Created:
    Updated:
    Resolved: