This is a storage backend using YDB for Jaeger
- collector/query: these are docker images with jaeger-collector/jaeger-query and plugin built-in into image
- watcher (see cmd/schema/watcher): this is schema-watcher that creates new tables for spans/indexes and removes old ones
cr.yandex/yc/jaeger-ydb-collector
cr.yandex/yc/jaeger-ydb-query
cr.yandex/yc/jaeger-ydb-watcher
cp docker-compose.example.yml docker-compose.yml
# edit docker-compose yml to set db dsn
docker-compose up -d
Name | Type | Default | Description |
---|---|---|---|
YDB_ADDRESS |
string |
db endpoint host:port to connect to | |
YDB_SECURE_CONNECTION |
string |
is secure connection enabled. One of ["enabled", "disabled"] | |
YDB_ANONYMOUS |
bool |
false |
anonymous auth |
YDB_TOKEN |
string |
auth token for internal purposes | |
YDB_SA_META_AUTH |
bool |
false |
use metadata to authorize requests (documentation) |
YDB_SA_ID |
string |
deprecated - use YDB_SA_KEY_JSON variable. service account id for Yandex.Cloud authorization (doc on service accounts: https://cloud.yandex.com/docs/iam/concepts/users/service-accounts) |
|
YDB_SA_KEY_ID |
string |
deprecated - use YDB_SA_KEY_JSON variable. service account key id for Yandex.Cloud authorization |
|
YDB_SA_PRIVATE_KEY_FILE |
string |
deprecated - use YDB_SA_KEY_JSON variable. to service account private key for Yandex.Cloud authorization |
|
YDB_SA_KEY_JSON |
string |
service account key for Yandex.Cloud authorization (doc on service accounts: https://cloud.yandex.com/docs/iam/concepts/users/service-accounts) in JSON . This variable replaces YDB_SA_ID , YDB_SA_KEY_ID and YDB_SA_PRIVATE_KEY_FILE |
|
YDB_PATH |
string |
database path | |
YDB_FOLDER |
string |
folder to store data in) | |
YDB_CONNECT_TIMEOUT |
duration |
10s |
db connect timeout |
YDB_WRITE_TIMEOUT |
duration |
write queries timeout | |
YDB_RETRY_ATTEMPT_TIMEOUT |
duration |
attempt to write queries timeout | |
YDB_READ_TIMEOUT |
duration |
10s |
read queries timeout |
YDB_READ_QUERY_PARALLEL |
integer |
16 |
controls number of parallel read subqueries |
YDB_READ_OP_LIMIT |
integer |
5000 |
max operation names to fetch for service |
YDB_READ_SVC_LIMIT |
integer |
1000 |
max service names to fetch |
YDB_POOL_SIZE |
integer |
100 |
db session pool size |
YDB_QUERY_CACHE_SIZE |
integer |
50 |
db query cache size |
YDB_WRITER_BUFFER_SIZE |
integer |
1000 |
span buffer size for batch writer |
YDB_WRITER_BATCH_SIZE |
integer |
100 |
number of spans in batch write calls |
YDB_WRITER_BATCH_WORKERS |
integer |
10 |
number of workers processing batch writes |
YDB_INDEXER_BUFFER_SIZE |
integer |
1000 |
span buffer size for indexer |
YDB_INDEXER_MAX_TRACES |
integer |
100 |
maximum trace_id count in a single index record |
YDB_INDEXER_MAX_TTL |
duration |
5s |
maximum amount of time for indexer to batch trace_ids for index records |
YDB_SCHEMA_NUM_PARTITIONS |
integer |
10 |
number of partitioned tables per day. Changing it requires recreating full data set |
Configuration options can be passed via config file. Use --grpc-storage-plugin.configuration-file
to pass configuration to YDB Plugin. In case of watcher use --config
for the same purpose.
Name | Type | Default | Description |
---|---|---|---|
WATCHER_AGE |
duration |
24h |
delete partition tables older than this value |
WATCHER_INTERVAL |
duration |
5m |
check interval |
YDB_FEATURE_SPLIT_BY_LOAD |
bool |
false |
enable table split by load feature |
YDB_FEATURE_COMPRESSION |
bool |
false |
enable table compression feature, used for span storage |