Dan Wang
Dan Wang
The number of metrics that are attached to server-level entity is the largest among all kinds of entities. They are distributed very widely on classes and source files. Thus server-level...
`neofinancial/[email protected]` always failed as below: ``` Run neofinancial/[email protected] with: token: *** titleFormat: %title% quiet: true bodyRegex: #(\d+) bodyURLRegex: http(s?):\/\/(github.com)(\/apache)(\/incubator-pegasus)(\/issues)\/\d+ ticketPrefix: # titleRegex: ^(CH)(-?)(\d{3,}) titleRegexFlags: gi branchRegex: ^(CH)(-?)(\d{3,}) branchRegexFlags: gi bodyRegexFlags:...
There are following metrics all of which are in class `replica`, [replica.cpp](https://github.com/apache/incubator-pegasus/blob/master/src/replica/replica.cpp) file: ``` _counters_table_level_latency[] ``` Therefore, table-level entity can just be defined in `replica.cpp`, where they are also instantiated...
As the new framework of metrics has nearly been completed (https://github.com/apache/incubator-pegasus/issues/922), it's time to migrate the metrics to the new framework. This issue is used as the main entrance, tracing...
The meta-related metrics migrated to new framework will be attached to *server* entity. All involved classes are put as below. Following metrics are the members of `server_state` ([server_state.cpp](https://github.com/apache/incubator-pegasus/blob/master/src/meta/server_state.cpp)), which is...
`nfs` metrics will be in use once `nfs_node` ([nfs_node.cpp](https://github.com/apache/incubator-pegasus/blob/master/src/nfs/nfs_node.cpp)) is created, which is a member of `replica_stub`. Thus in new framework all of them will be attached to *server* entity....
Some classes that have few of replica-level metrics are collected into one issue to be migrated. Following metrics are the members of `pegasus_event_listener` ([pegasus_event_listener.cpp](https://github.com/apache/incubator-pegasus/blob/master/src/server/pegasus_event_listener.cpp)), which is created at the construction...
This PR is to resolve https://github.com/apache/incubator-pegasus/issues/1010.
In jemalloc, there are 2 important options for the two-phase, decay-based purging, with each option corresponding to a phase. They are: - [dirty_decay_ms](http://jemalloc.net/jemalloc.3.html#arena.i.dirty_decay_ms) (from state `dirty` to `muzzy`) - [muzzy_decay_ms](http://jemalloc.net/jemalloc.3.html#arena.i.muzzy_decay_ms)...
The block size used to read the files of mutation logs should be configurable to limit the memory consumed by rDSN, with `log_private_block_bytes` specified for private logs and `log_shared_block_bytes` for...