[
https://issues.apache.org/jira/browse/IGNITE-3619?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ksenia Rybakova updated IGNITE-3619:
------------------------------------
Description:
The following warnings appear during load test:
{noformat}
[15:46:07,413][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running transaction [startTime=15:45:04.471, curTime=15:46:07.356,
tx=GridNearTxLocal [mappings=IgniteTxMappingsImpl [], nearLocallyMapped
...
[15:46:07,787][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running cache future [startTime=15:45:04.775, curTime=15:46:07.356,
fut=GridNearTxFinishFuture [futId=905d64b4651-3154c7e5-c2bd-46d5-a2ae-
...
[15:46:07,830][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running cache operations, dump threads.
[15:46:07,831][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] No
deadlocked threads detected.
{noformat}
In some cases these warnings appear during preloading, in other cases during
performing main test.
Sometimes CPU at part of the nodes drops almost to 0% and test hung.
Configs:
1)
- 5 hosts
- benchmark: CacheRandomOperation
- Operations: all (put, put_all, get, get_all, invoke, invoke_all, remove,
remove_all, put_if_absent, replace, scan_query, sql_query, continuous_query)
- Heap: 2Gb for servers, 1Gb for clients
- Number of clients to number of servers: 1:1 (48 clients, 48 servers)
- Number of caches: 12
- Types of caches (atomicity mode): atomic
- Types of caches (tiered storage mode): different (onheap without eviction,
onheap with eviction, offheap_tired, offheap_values)
- Types of caches (indexing): different (with and without indexes)
- Types of caches (cache mode): partitioned
- Backups count: 1
- Duration: 1h
2)
- 5 hosts
- benchmark: CacheRandomOperation
- Operations: remove, replace
- Heap: 4Gb for servers, 1Gb for clients
- Number of clients to number of servers: 4:1 (80 clients, 20 servers)
- Number of caches: 12
- Types of caches (atomicity mode): different (atomic, transactional)
- Types of caches (tiered storage mode): different (onheap without eviction,
onheap with eviction, offheap_tired, offheap_values)
- Types of caches (indexing): different (with and without indexes)
- Types of caches (cache mode): partitioned
- Backups count: 1
- Duration: 1h
Property files and xml configs are attached.
Logs can be provided on demand.
was:
The following warnings appear during load test:
{noformat}
[15:46:07,413][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running transaction [startTime=15:45:04.471, curTime=15:46:07.356,
tx=GridNearTxLocal [mappings=IgniteTxMappingsImpl [], nearLocallyMapped
...
[15:46:07,787][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running cache future [startTime=15:45:04.775, curTime=15:46:07.356,
fut=GridNearTxFinishFuture [futId=905d64b4651-3154c7e5-c2bd-46d5-a2ae-
...
[15:46:07,830][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found long
running cache operations, dump threads.
[15:46:07,831][WARN
][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] No
deadlocked threads detected.
{noformat}
In some cases these warnings appear during preloading, in other cases during
performing main test.
Sometimes CPU at part of the nodes drops almost to 0% and test hung.
Configs:
1)
- 5 hosts
- benchmark: CacheRandomOperation
- Operations: all (put, put_all, get, get_all, invoke, invoke_all, remove,
remove_all, put_if_absent, replace, scan_query, sql_query, continuous_query)
- Heap: 2Gb for servers, 1Gb for clients
- Number of clients to number of servers: 1:1 (48 clients, 48 servers)
- Number of caches: 12
- Types of caches (atomicity mode): atomic
- Types of caches (tiered storage mode): different (onheap without eviction,
onheap with eviction, offheap_tired, offheap_values)
- Types of caches (indexing): different (with and without indexes)
- Types of caches (cache mode): partitioned
- Backups count: 1
- Duration: 1h
2)
- 5 hosts
- benchmark: CacheRandomOperation
- Operations: all (put, put_all, get, get_all, invoke, invoke_all, remove,
remove_all, put_if_absent, replace, scan_query, sql_query, continuous_query)
- Heap: 8Gb for servers, 4Gb for clients
- Number of clients to number of servers: 1:4 (5 clients, 20 servers)
- Number of caches: 12
- Types of caches (atomicity mode): transactional
- Types of caches (tiered storage mode): different (onheap without eviction,
onheap with eviction, offheap_tired, offheap_values)
- Types of caches (indexing): different (with and without indexes)
- Types of caches (cache mode): partitioned
- Backups count: 1
- Duration: 1h
Property files and xml configs are attached.
Logs can be provided on demand.
> "long running transaction/cache operations" during load tests
> -------------------------------------------------------------
>
> Key: IGNITE-3619
> URL: https://issues.apache.org/jira/browse/IGNITE-3619
> Project: Ignite
> Issue Type: Bug
> Affects Versions: 1.6
> Reporter: Ksenia Rybakova
> Attachments: ignite-base-load-config.xml_1,
> ignite-base-load-config.xml_2, run-load.properties_1, run-load.properties_2,
> run-load.xml_1, run-load.xml_2
>
>
> The following warnings appear during load test:
> {noformat}
> [15:46:07,413][WARN
> ][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found
> long running transaction [startTime=15:45:04.471, curTime=15:46:07.356,
> tx=GridNearTxLocal [mappings=IgniteTxMappingsImpl [], nearLocallyMapped
> ...
> [15:46:07,787][WARN
> ][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found
> long running cache future [startTime=15:45:04.775, curTime=15:46:07.356,
> fut=GridNearTxFinishFuture [futId=905d64b4651-3154c7e5-c2bd-46d5-a2ae-
> ...
> [15:46:07,830][WARN
> ][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] Found
> long running cache operations, dump threads.
> [15:46:07,831][WARN
> ][grid-timeout-worker-#79%null%][GridCachePartitionExchangeManager] No
> deadlocked threads detected.
> {noformat}
> In some cases these warnings appear during preloading, in other cases during
> performing main test.
> Sometimes CPU at part of the nodes drops almost to 0% and test hung.
> Configs:
> 1)
> - 5 hosts
> - benchmark: CacheRandomOperation
> - Operations: all (put, put_all, get, get_all, invoke, invoke_all, remove,
> remove_all, put_if_absent, replace, scan_query, sql_query, continuous_query)
> - Heap: 2Gb for servers, 1Gb for clients
> - Number of clients to number of servers: 1:1 (48 clients, 48 servers)
> - Number of caches: 12
> - Types of caches (atomicity mode): atomic
> - Types of caches (tiered storage mode): different (onheap without eviction,
> onheap with eviction, offheap_tired, offheap_values)
> - Types of caches (indexing): different (with and without indexes)
> - Types of caches (cache mode): partitioned
> - Backups count: 1
> - Duration: 1h
> 2)
> - 5 hosts
> - benchmark: CacheRandomOperation
> - Operations: remove, replace
> - Heap: 4Gb for servers, 1Gb for clients
> - Number of clients to number of servers: 4:1 (80 clients, 20 servers)
> - Number of caches: 12
> - Types of caches (atomicity mode): different (atomic, transactional)
> - Types of caches (tiered storage mode): different (onheap without eviction,
> onheap with eviction, offheap_tired, offheap_values)
> - Types of caches (indexing): different (with and without indexes)
> - Types of caches (cache mode): partitioned
> - Backups count: 1
> - Duration: 1h
> Property files and xml configs are attached.
> Logs can be provided on demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
