Details
-
Bug
-
Resolution: Fixed
-
Critical
-
7.6.2
-
7.6.2 build 3703
-
Untriaged
-
-
0
-
Unknown
Description
Test environment:
- 5 dedicated fts nodes with 8cpus and 16GB RAM each node
- 60Million document on KV of dimension 1536
Test Steps:
- Loaded ~60Mil documents on to KV
- Created indexing with 18 partitions per node, so overall 90 paritions
- "planParams": { "maxPartitionsPerPIndex": 12, "indexPartitions": 90 },
- I have only one index with l2_norm
- Wait for indexing complete and cluster to be quiet.
- Ran 5 knn queries with k=10 and fts exited with status 2
ns_1@172.23.97.108 |
3:00:07 AM 11 Jun, 2024 |
|
Service 'fts' exited with status 2. Restarting. Messages: |
goroutine 1577357 gp=0xc04e42ba40 m=nil [runnable]: |
github.com/blevesearch/bleve/v2/index/scorch.(*OptimizeVR).Finish.func1.gowrap2()
|
/home/couchbase/.cbdepscache/gomodcache/pkg/mod/github.com/blevesearch/bleve/v2@v2.4.1-0.20240606135638-72887d93ad69/index/scorch/optimize_knn.go:110 fp=0xc03d45b7e0 sp=0xc03d45b7d8 pc=0x78be40 |
runtime.goexit({})
|
/home/couchbase/.cbdepscache/exploded/x86_64/go-1.22.2/go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc03d45b7e8 sp=0xc03d45b7e0 pc=0x47cd01 |
created by github.com/blevesearch/bleve/v2/index/scorch.(*OptimizeVR).Finish.func1 in goroutine 1576227 |
/home/couchbase/.cbdepscache/gomodcache/pkg/mod/github.com/blevesearch/bleve/v2@v2.4.1-0.20240606135638-72887d93ad69/index/scorch/optimize_knn.go:110 +0x115rax 0x0 |
rbx 0x6 |
rcx 0x7fa6ff0367bb |
rdx 0x0 |
rdi 0x2 |
rsi 0x7fa678bc1e80 |
rbp 0x2008952 |
rsp 0x7fa678bc1e80 |
r8 0x0 |
r9 0x7fa678bc1e80 |
r10 0x8 |
r11 0x246 |
r12 0x7fa678bc2290 |
r13 0x599010 |
r14 0xc000006c40 |
r15 0xfffffffffffff |
rip 0x7fa6ff0367bb |
rflags 0x246 |
cs 0x33 |
fs 0x0 |
gs 0x0 |
Image indicating just ~4 queries:
Also I see total_knn_queries in promtimer stat as 305 which is contradictory, there seems to be some discrepancy in this stat and I also see all 305 being marked as slow.