Memcached
Memcached collector can collect the running status metrics from Memcached instances, and collect the metrics to the Guance to help monitor and analyze various abnormal situations of Memcached.
Config¶
Preconditions¶
- Memcached version >=
1.5.0. Already tested version:- 1.5.x
- 1.6.x
Collector Configuration¶
Go to the conf.d/db directory under the DataKit installation directory, copy memcached.conf.sample and name it memcached.conf. Examples are as follows:
[[inputs.memcached]]
## Servers' addresses.
servers = ["localhost:11211"]
# unix_sockets = ["/var/run/memcached.sock"]
## Set true to enable election
election = true
## Collect extra stats
# extra_stats = ["slabs", "items"]
## Collect interval.
# units: "ns", "us", "ms", "s", "m", "h"
interval = "10s"
[inputs.memcached.tags]
# some_tag = "some_value"
# more_tag = "some_other_value"
# ...
Once configured, restart DataKit.
The collector can now be turned on by ConfigMap Injection Collector Configuration.
Metric¶
For all of the following data collections, the global election tags will added automatically, we can add extra tags in [inputs.memcached.tags] if needed:
memcached¶
- Tags
| Tag | Description |
|---|---|
| server | The host name from which metrics are gathered |
- Metrics
| Metric | Description |
|---|---|
| accepting_conns | Whether or not server is accepting conns Type: int Unit: count |
| auth_cmds | Number of authentication commands handled, success or failure Type: int Unit: count |
| auth_errors | Number of failed authentications Type: int Unit: count |
| bytes | Current number of bytes used to store items Type: int Unit: digital,B |
| bytes_read | Total number of bytes read by this server from network Type: int Unit: digital,B |
| bytes_written | Total number of bytes sent by this server to network Type: int Unit: digital,B |
| cas_badval | Number of CAS for which a key was found, but the CAS value did not match Type: int Unit: count |
| cas_hits | Number of successful CAS requests Type: int Unit: count |
| cas_misses | Number of CAS requests against missing keys Type: int Unit: count |
| cmd_flush | Cumulative number of flush requests Type: int Unit: count |
| cmd_get | Cumulative number of retrieval requests Type: int Unit: count |
| cmd_set | Cumulative number of storage requests Type: int Unit: count |
| cmd_touch | Cumulative number of touch requests Type: int Unit: count |
| conn_yields | Number of times any connection yielded to another due to hitting the -R limit Type: int Unit: count |
| connection_structures | Number of connection structures allocated by the server Type: int Unit: count |
| curr_connections | Number of open connections Type: int Unit: count |
| curr_items | Current number of items stored Type: int Unit: count |
| decr_hits | Number of successful decr requestsType: int Unit: count |
| decr_misses | Number of decr requests against missing keysType: int Unit: count |
| delete_hits | Number of deletion requests resulting in an item being removed Type: int Unit: count |
| delete_misses | umber of deletions requests for missing keys Type: int Unit: count |
| evicted_unfetched | Items evicted from LRU that were never touched by get/incr/append/etc Type: int Unit: count |
| evictions | Number of valid items removed from cache to free memory for new items Type: int Unit: count |
| expired_unfetched | Items pulled from LRU that were never touched by get/incr/append/etc before expiring Type: int Unit: count |
| get_hits | Number of keys that have been requested and found present Type: int Unit: count |
| get_misses | Number of items that have been requested and not found Type: int Unit: count |
| hash_bytes | Bytes currently used by hash tables Type: int Unit: digital,B |
| hash_is_expanding | Indicates if the hash table is being grown to a new size Type: int Unit: count |
| hash_power_level | Current size multiplier for hash table Type: int Unit: count |
| incr_hits | Number of successful incr requests Type: int Unit: count |
| incr_misses | Number of incr requests against missing keys Type: int Unit: count |
| limit_maxbytes | Number of bytes this server is allowed to use for storage Type: int Unit: digital,B |
| listen_disabled_num | Number of times server has stopped accepting new connections (maxconns)Type: int Unit: count |
| reclaimed | Number of times an entry was stored using memory from an expired entry Type: int Unit: count |
| threads | Number of worker threads requested Type: int Unit: count |
| total_connections | Total number of connections opened since the server started running Type: int Unit: count |
| total_items | Total number of items stored since the server started Type: int Unit: count |
| touch_hits | Number of keys that have been touched with a new expiration time Type: int Unit: count |
| touch_misses | Number of items that have been touched and not found Type: int Unit: count |
| uptime | Number of secs since the server started Type: int Unit: count |
memcached_items¶
- Tags
| Tag | Description |
|---|---|
| server | The host name from which metrics are gathered |
| slab_id | The id of the current slab |
- Metrics
| Metric | Description |
|---|---|
| age | Age of the oldest item in the LRU Type: int Unit: count |
| crawler_reclaimed | Number of items which freed by the LRU Crawler Type: int Unit: count |
| direct_reclaims | Number of worker threads which had to directly pull LRU tails to find memory for a new item Type: int Unit: count |
| evicted | Number of the items which had to be evicted from the LRU before expiring Type: int Unit: count |
| evicted_nonzero | Number of the onzero items which had an explicit expire time set had to be evicted from the LRU before expiringType: int Unit: count |
| evicted_time | Seconds since the last access for the most recent item evicted from this class Type: int Unit: time,s |
| evicted_unfetched | Number of the valid items evicted from the LRU which were never touched after being set Type: int Unit: count |
| expired_unfetched | Number of the expired items reclaimed from the LRU which were never touched after being set Type: int Unit: count |
| lrutail_reflocked | Number of items which found to be refcount locked in the LRU tailType: int Unit: count |
| moves_to_cold | Number of items which were moved from HOT or WARM into COLD Type: int Unit: count |
| moves_to_warm | Number of items which were moved from COLD to WARM Type: int Unit: count |
| moves_within_lru | Number of active items which were bumped within HOT or WARM Type: int Unit: count |
| number | Number of items presently stored in this slab class Type: int Unit: count |
| number_cold | Number of items presently stored in the COLD LRU Type: int Unit: count |
| number_hot | Number of items presently stored in the HOT LRU Type: int Unit: count |
| number_noexp | Number of items presently stored in the NOEXP classType: int Unit: count |
| number_warm | Number of items presently stored in the WARM LRU Type: int Unit: count |
| outofmemory | Number of the underlying slab class which was unable to store a new item Type: int Unit: count |
| reclaimed | Number of entries which were stored using memory from an expired entry Type: int Unit: count |
| tailrepairs | How many times memcache self-healed a slab with a refcount leakType: int Unit: count |
memcached_slabs¶
- Tags
| Tag | Description |
|---|---|
| server | The host name from which metrics are gathered |
| slab_id | The id of the current slab |
- Metrics
| Metric | Description |
|---|---|
| active_slabs | Total number of slab classes allocated Type: int Unit: count |
| cas_badval | Number of CAS commands failed to modify a value due to a bad CAS id Type: int Unit: count |
| cas_hits | Number of CAS commands modified this slab class Type: int Unit: count |
| chunk_size | The amount of space each chunk uses Type: int Unit: digital,B |
| chunks_per_page | How many chunks exist within one page Type: int Unit: count |
| cmd_set | Number of set requests stored data in this slab class Type: int Unit: count |
| decr_hits | Number of decrs commands modified this slab classType: int Unit: count |
| delete_hits | Number of delete commands succeeded in this slab class Type: int Unit: count |
| free_chunks | Chunks not yet allocated to items or freed via delete Type: int Unit: count |
| free_chunks_end | Number of free chunks at the end of the last allocated page Type: int Unit: count |
| get_hits | Number of get requests were serviced by this slab class Type: int Unit: count |
| incr_hits | Number of incrs commands modified this slab classType: int Unit: count |
| total_chunks | Total number of chunks allocated to the slab class Type: int Unit: count |
| total_malloced | Total amount of memory allocated to slab pages Type: int Unit: digital,B |
| total_pages | Total number of pages allocated to the slab class Type: int Unit: count |
| touch_hits | Number of touches serviced by this slab class Type: int Unit: count |
| used_chunks | How many chunks have been allocated to items Type: int Unit: count |