1VARNISH-COUNTERS(7)                                        VARNISH-COUNTERS(7)
2
3
4

NAME

6       varnish-counters - Varnish counter field definitions
7

MAIN – MAIN COUNTERS

9       summs counter - debug
10          stat summ operations
11
12          Number  of  times  per-thread statistics were summed into the global
13          counters.
14
15       uptime counter - info
16          Child process uptime
17
18          How long the child process has been running.
19
20       sess_conn counter - info
21          Sessions accepted
22
23          Count of sessions successfully accepted
24
25       sess_fail counter - info
26          Session accept failures
27
28          Count of failures to accept TCP connection.
29
30          This counter is the sum of the sess_fail_* counters, which give more
31          detailed information.
32
33       sess_fail_econnaborted counter - info
34          Session accept failures: connection aborted
35
36          Detailed  reason  for  sess_fail:  Connection aborted by the client,
37          usually harmless.
38
39       sess_fail_eintr counter - info
40          Session accept failures: interrupted system call
41
42          Detailed reason for sess_fail: The accept()  call  was  interrupted,
43          usually harmless
44
45       sess_fail_emfile counter - info
46          Session accept failures: too many open files
47
48          Detailed  reason  for  sess_fail:  No file descriptor was available.
49          Consider raising RLIMIT_NOFILE (see ulimit -n).
50
51       sess_fail_ebadf counter - info
52          Session accept failures: bad file descriptor
53
54          Detailed reason for sess_fail: The listen socket file descriptor was
55          invalid. Should never happen.
56
57       sess_fail_enomem counter - info
58          Session accept failures: not enough memory
59
60          Detailed  reason for sess_fail: Most likely insufficient socket buf‐
61          fer memory. Should never happen
62
63       sess_fail_other counter - info
64          Session accept failures: other
65
66          Detailed reason for sess_fail: neither of the above,  see  SessError
67          log (varnishlog -g raw -i SessError).
68
69       client_req_400 counter - info
70          Client requests received, subject to 400 errors
71
72          400 means we couldn't make sense of the request, it was malformed in
73          some drastic way.
74
75       client_req_417 counter - info
76          Client requests received, subject to 417 errors
77
78          417 means that something went wrong with an Expect: header.
79
80       client_req counter - info
81          Good client requests received
82
83          The count of parseable client requests seen.
84
85       esi_req counter - info
86          ESI subrequests
87
88          Number of ESI subrequests made.
89
90       cache_hit counter - info
91          Cache hits
92
93          Count of cache hits.  A cache hit indicates that an object has  been
94          delivered to a client without fetching it from a backend server.
95
96       cache_hit_grace counter - info
97          Cache grace hits
98
99          Count  of  cache  hits with grace. A cache hit with grace is a cache
100          hit where the object is expired. Note that such hits  are  also  in‐
101          cluded in the cache_hit counter.
102
103       cache_hitpass counter - info
104          Cache hits for pass.
105
106          Count  of hits for pass. A cache hit for pass indicates that Varnish
107          is going to pass the request to the backend and  this  decision  has
108          been  cached in it self. This counts how many times the cached deci‐
109          sion is being used.
110
111       cache_hitmiss counter - info
112          Cache hits for miss.
113
114          Count of hits for miss. A cache hit for miss indicates that  Varnish
115          is  going to proceed as for a cache miss without request coalescing,
116          and this decision has been cached. This counts how  many  times  the
117          cached decision is being used.
118
119       cache_miss counter - info
120          Cache misses
121
122          Count  of misses. A cache miss indicates the object was fetched from
123          the backend before delivering it to the client.
124
125       beresp_uncacheable counter - info
126              Uncacheable backend responses
127
128          Count of backend responses considered uncacheable.
129
130       beresp_shortlived counter - info
131              Shortlived objects
132
133          Count of  objects  created  with  ttl+grace+keep  shorter  than  the
134          'shortlived' runtime parameter.
135
136       backend_conn counter - info
137          Backend conn. success
138
139          How many backend connections have successfully been established.
140
141       backend_unhealthy counter - info
142          Backend conn. not attempted
143
144       backend_busy counter - info
145          Backend conn. too many
146
147       backend_fail counter - info
148          Backend conn. failures
149
150       backend_reuse counter - info
151          Backend conn. reuses
152
153          Count  of backend connection reuses. This counter is increased when‐
154          ever we reuse a recycled connection.
155
156       backend_recycle counter - info
157          Backend conn. recycles
158
159          Count of backend connection  recycles.  This  counter  is  increased
160          whenever  we  have a keep-alive connection that is put back into the
161          pool of connections. It has not yet been used, but it might be,  un‐
162          less the backend closes it.
163
164       backend_retry counter - info
165          Backend conn. retry
166
167       fetch_head counter - info
168          Fetch no body (HEAD)
169
170          beresp with no body because the request is HEAD.
171
172       fetch_length counter - info
173          Fetch with Length
174
175          beresp.body with Content-Length.
176
177       fetch_chunked counter - info
178          Fetch chunked
179
180          beresp.body with Chunked.
181
182       fetch_eof counter - info
183          Fetch EOF
184
185          beresp.body with EOF.
186
187       fetch_bad counter - info
188          Fetch bad T-E
189
190          beresp.body length/fetch could not be determined.
191
192       fetch_none counter - info
193          Fetch no body
194
195          beresp.body empty
196
197       fetch_1xx counter - info
198          Fetch no body (1xx)
199
200          beresp with no body because of 1XX response.
201
202       fetch_204 counter - info
203          Fetch no body (204)
204
205          beresp with no body because of 204 response.
206
207       fetch_304 counter - info
208          Fetch no body (304)
209
210          beresp with no body because of 304 response.
211
212       fetch_failed counter - info
213          Fetch failed (all causes)
214
215          beresp fetch failed.
216
217       fetch_no_thread counter - info
218          Fetch failed (no thread)
219
220          beresp fetch failed, no thread available.
221
222       pools gauge - info
223          Number of thread pools
224
225          Number  of  thread pools. See also parameter thread_pools. NB: Pres‐
226          ently pools cannot be removed once created.
227
228       threads gauge - info
229          Total number of threads
230
231          Number of threads in all pools. See  also  parameters  thread_pools,
232          thread_pool_min and thread_pool_max.
233
234       threads_limited counter - info
235          Threads hit max
236
237          Number of times more threads were needed, but limit was reached in a
238          thread pool. See also parameter thread_pool_max.
239
240       threads_created counter - info
241          Threads created
242
243          Total number of threads created in all pools.
244
245       threads_destroyed counter - info
246          Threads destroyed
247
248          Total number of threads destroyed in all pools.
249
250       threads_failed counter - info
251          Thread creation failed
252
253          Number of times creating a thread failed. See VSL::Debug  for  diag‐
254          nostics. See also parameter thread_fail_delay.
255
256       thread_queue_len gauge - info
257          Length of session queue
258
259          Length  of  session queue waiting for threads. NB: Only updates once
260          per second. See also parameter thread_queue_limit.
261
262       busy_sleep counter - info
263          Number of requests sent to sleep on busy objhdr
264
265          Number of requests sent to sleep without  a  worker  thread  because
266          they found a busy object.
267
268       busy_wakeup counter - info
269          Number of requests woken after sleep on busy objhdr
270
271          Number of requests taken off the busy object sleep list and resched‐
272          uled.
273
274       busy_killed counter - info
275          Number of requests killed after sleep on busy objhdr
276
277          Number of requests killed from the busy object  sleep  list  due  to
278          lack of resources.
279
280       sess_queued counter - info
281          Sessions queued for thread
282
283          Number  of  times  session was queued waiting for a thread. See also
284          parameter thread_queue_limit.
285
286       sess_dropped counter - info
287          Sessions dropped for thread
288
289          Number of times an HTTP/1 session was dropped because the queue  was
290          too long already. See also parameter thread_queue_limit.
291
292       req_dropped counter - info
293          Requests dropped
294
295          Number  of  times an HTTP/2 stream was refused because the queue was
296          too long already. See also parameter thread_queue_limit.
297
298       n_object gauge - info
299          object structs made
300
301          Approximate number of HTTP objects (headers + body, if  present)  in
302          the cache.
303
304       n_vampireobject gauge - diag
305          unresurrected objects
306
307          Number of unresurrected objects
308
309       n_objectcore gauge - info
310          objectcore structs made
311
312          Approximate  number  of  object metadata elements in the cache. Each
313          object needs an objectcore, extra objectcores are for  hit-for-miss,
314          hit-for-pass and busy objects.
315
316       n_objecthead gauge - info
317          objecthead structs made
318
319          Approximate number of different hash entries in the cache.
320
321       n_backend gauge - info
322          Number of backends
323
324          Number of backends known to us.
325
326       n_expired counter - info
327          Number of expired objects
328
329          Number of objects that expired from cache because of old age.
330
331       n_lru_nuked counter - info
332          Number of LRU nuked objects
333
334          How  many  objects have been forcefully evicted from storage to make
335          room for a new object.
336
337       n_lru_moved counter - diag
338          Number of LRU moved objects
339
340          Number of move operations done on the LRU list.
341
342       n_lru_limited counter - info
343          Reached nuke_limit
344
345          Number of times more  storage  space  were  needed,  but  limit  was
346          reached in a nuke_limit. See also parameter nuke_limit.
347
348       losthdr counter - info
349          HTTP header overflows
350
351       s_sess counter - info
352          Total sessions seen
353
354       n_pipe gauge - info
355          Number of ongoing pipe sessions
356
357       pipe_limited counter - info
358          Pipes hit pipe_sess_max
359
360          Number  of  times more pipes were needed, but the limit was reached.
361          See also parameter pipe_sess_max.
362
363       s_pipe counter - info
364          Total pipe sessions seen
365
366       s_pass counter - info
367          Total pass-ed requests seen
368
369       s_fetch counter - info
370          Total backend fetches initiated
371
372          Total backend fetches initiated, including background fetches.
373
374       s_bgfetch counter - info
375          Total backend background fetches initiated
376
377       s_synth counter - info
378          Total synthetic responses made
379
380       s_req_hdrbytes counter - info
381          Request header bytes
382
383          Total request header bytes received
384
385       s_req_bodybytes counter - info
386          Request body bytes
387
388          Total request body bytes received
389
390       s_resp_hdrbytes counter - info
391          Response header bytes
392
393          Total response header bytes transmitted
394
395       s_resp_bodybytes counter - info
396          Response body bytes
397
398          Total response body bytes transmitted
399
400       s_pipe_hdrbytes counter - info
401          Pipe request header bytes
402
403          Total request bytes received for piped sessions
404
405       s_pipe_in counter - info
406          Piped bytes from client
407
408          Total number of bytes forwarded from clients in pipe sessions
409
410       s_pipe_out counter - info
411          Piped bytes to client
412
413          Total number of bytes forwarded to clients in pipe sessions
414
415       sess_closed counter - info
416          Session Closed
417
418       sess_closed_err counter - info
419          Session Closed with error
420
421          Total number of sessions closed with errors. See sc_* diag  counters
422          for detailed breakdown
423
424       sess_readahead counter - info
425          Session Read Ahead
426
427       sess_herd counter - diag
428          Session herd
429
430          Number of times the timeout_linger triggered
431
432       sc_rem_close counter - diag
433          Session OK  REM_CLOSE
434
435          Number of session closes with REM_CLOSE (Client Closed)
436
437       sc_req_close counter - diag
438          Session OK  REQ_CLOSE
439
440          Number of session closes with REQ_CLOSE (Client requested close)
441
442       sc_req_http10 counter - diag
443          Session Err REQ_HTTP10
444
445          Number of session closes with Error REQ_HTTP10 (Proto < HTTP/1.1)
446
447       sc_rx_bad counter - diag
448          Session Err RX_BAD
449
450          Number of session closes with Error RX_BAD (Received bad req/resp)
451
452       sc_rx_body counter - diag
453          Session Err RX_BODY
454
455          Number  of  session  closes  with  Error  RX_BODY (Failure receiving
456          req.body)
457
458       sc_rx_junk counter - diag
459          Session Err RX_JUNK
460
461          Number of session closes with Error RX_JUNK (Received junk data)
462
463       sc_rx_overflow counter - diag
464          Session Err RX_OVERFLOW
465
466          Number of session closes with  Error  RX_OVERFLOW  (Received  buffer
467          overflow)
468
469       sc_rx_timeout counter - diag
470          Session Err RX_TIMEOUT
471
472          Number of session closes with Error RX_TIMEOUT (Receive timeout)
473
474       sc_rx_close_idle counter - diag
475          Session Err RX_CLOSE_IDLE
476
477          Number  of session closes with Error RX_CLOSE_IDLE: timeout_idle has
478          been exceeded while waiting for a client request.
479
480       sc_tx_pipe counter - diag
481          Session OK  TX_PIPE
482
483          Number of session closes with TX_PIPE (Piped transaction)
484
485       sc_tx_error counter - diag
486          Session Err TX_ERROR
487
488          Number of session closes with Error TX_ERROR (Error transaction)
489
490       sc_tx_eof counter - diag
491          Session OK  TX_EOF
492
493          Number of session closes with TX_EOF (EOF transmission)
494
495       sc_resp_close counter - diag
496          Session OK  RESP_CLOSE
497
498          Number of session  closes  with  RESP_CLOSE  (Backend/VCL  requested
499          close)
500
501       sc_overload counter - diag
502          Session Err OVERLOAD
503
504          Number of session closes with Error OVERLOAD (Out of some resource)
505
506       sc_pipe_overflow counter - diag
507          Session Err PIPE_OVERFLOW
508
509          Number  of  session  closes  with  Error PIPE_OVERFLOW (Session pipe
510          overflow)
511
512       sc_range_short counter - diag
513          Session Err RANGE_SHORT
514
515          Number of session closes with Error RANGE_SHORT  (Insufficient  data
516          for range)
517
518       sc_req_http20 counter - diag
519          Session Err REQ_HTTP20
520
521          Number of session closes with Error REQ_HTTP20 (HTTP2 not accepted)
522
523       sc_vcl_failure counter - diag
524          Session Err VCL_FAILURE
525
526          Number of session closes with Error VCL_FAILURE (VCL failure)
527
528       client_resp_500 counter - diag
529          Delivery failed due to insufficient workspace.
530
531          Number of times we failed a response due to running out of workspace
532          memory during delivery.
533
534       ws_backend_overflow counter - diag
535          workspace_backend overflows
536
537          Number of times we ran out of space in workspace_backend.
538
539       ws_client_overflow counter - diag
540          workspace_client overflows
541
542          Number of times we ran out of space in workspace_client.
543
544       ws_thread_overflow counter - diag
545          workspace_thread overflows
546
547          Number of times we ran out of space in workspace_thread.
548
549       ws_session_overflow counter - diag
550          workspace_session overflows
551
552          Number of times we ran out of space in workspace_session.
553
554       shm_records counter - diag
555          SHM records
556
557       shm_writes counter - diag
558          SHM writes
559
560       shm_flushes counter - diag
561          SHM flushes due to overflow
562
563       shm_cont counter - diag
564          SHM MTX contention
565
566       shm_cycles counter - diag
567          SHM cycles through buffer
568
569       backend_req counter - info
570          Backend requests made
571
572       n_vcl gauge - info
573          Number of loaded VCLs in total
574
575       n_vcl_avail gauge - diag
576          Number of VCLs available
577
578       n_vcl_discard gauge - diag
579          Number of discarded VCLs
580
581       vcl_fail counter - info
582          VCL failures
583
584          Count of failures which prevented VCL from completing.
585
586       bans gauge - info
587          Count of bans
588
589          Number of all bans in system, including  bans  superseded  by  newer
590          bans and bans already checked by the ban-lurker.
591
592       bans_completed gauge - diag
593          Number of bans marked 'completed'
594
595          Number  of  bans which are no longer active, either because they got
596          checked by the ban-lurker or superseded by newer identical bans.
597
598       bans_obj gauge - diag
599          Number of bans using obj.*
600
601          Number of bans which use obj.* variables.  These bans  can  possibly
602          be washed by the ban-lurker.
603
604       bans_req gauge - diag
605          Number of bans using req.*
606
607          Number  of  bans  which  use req.* variables.  These bans can not be
608          washed by the ban-lurker.
609
610       bans_added counter - diag
611          Bans added
612
613          Counter of bans added to ban list.
614
615       bans_deleted counter - diag
616          Bans deleted
617
618          Counter of bans deleted from ban list.
619
620       bans_tested counter - diag
621          Bans tested against objects (lookup)
622
623          Count of how many bans and objects have  been  tested  against  each
624          other during hash lookup.
625
626       bans_obj_killed counter - diag
627          Objects killed by bans (lookup)
628
629          Number of objects killed by bans during object lookup.
630
631       bans_lurker_tested counter - diag
632          Bans tested against objects (lurker)
633
634          Count  of  how  many  bans and objects have been tested against each
635          other by the ban-lurker.
636
637       bans_tests_tested counter - diag
638          Ban tests tested against objects (lookup)
639
640          Count of how many tests and objects have been  tested  against  each
641          other  during  lookup.  'ban req.url == foo && req.http.host == bar'
642          counts as one in 'bans_tested' and as two in 'bans_tests_tested'
643
644       bans_lurker_tests_tested counter - diag
645          Ban tests tested against objects (lurker)
646
647          Count of how many tests and objects have been  tested  against  each
648          other  by  the  ban-lurker.  'ban req.url == foo && req.http.host ==
649          bar'   counts   as   one   in   'bans_tested'   and   as   two    in
650          'bans_tests_tested'
651
652       bans_lurker_obj_killed counter - diag
653          Objects killed by bans (lurker)
654
655          Number of objects killed by the ban-lurker.
656
657       bans_lurker_obj_killed_cutoff counter - diag
658          Objects killed by bans for cutoff (lurker)
659
660          Number  of  objects  killed  by the ban-lurker to keep the number of
661          bans below ban_cutoff.
662
663       bans_dups counter - diag
664          Bans superseded by other bans
665
666          Count of bans replaced by later identical bans.
667
668       bans_lurker_contention counter - diag
669          Lurker gave way for lookup
670
671          Number of times the ban-lurker had to wait for lookups.
672
673       bans_persisted_bytes gauge - diag
674          Bytes used by the persisted ban lists
675
676          Number of bytes used by the persisted ban lists.
677
678       bans_persisted_fragmentation gauge - diag
679          Extra bytes in persisted ban lists due to fragmentation
680
681          Number of extra bytes accumulated through dropped and completed bans
682          in the persistent ban lists.
683
684       n_purges counter - info
685          Number of purge operations executed
686
687       n_obj_purged counter - info
688          Number of purged objects
689
690       exp_mailed counter - diag
691          Number of objects mailed to expiry thread
692
693          Number of objects mailed to expiry thread for handling.
694
695       exp_received counter - diag
696          Number of objects received by expiry thread
697
698          Number of objects received by expiry thread for handling.
699
700       hcb_nolock counter - debug
701          HCB Lookups without lock
702
703       hcb_lock counter - debug
704          HCB Lookups with lock
705
706       hcb_insert counter - debug
707          HCB Inserts
708
709       esi_errors counter - diag
710          ESI parse errors (unlock)
711
712       esi_warnings counter - diag
713          ESI parse warnings (unlock)
714
715       vmods gauge - info
716          Loaded VMODs
717
718       n_gzip counter - info
719          Gzip operations
720
721       n_gunzip counter - info
722          Gunzip operations
723
724       n_test_gunzip counter - info
725          Test gunzip operations
726
727          Those  operations  occur  when  Varnish receives a compressed object
728          from a backend. They are done to verify the gzip stream  while  it's
729          inserted in storage.
730

MGT – MANAGEMENT PROCESS COUNTERS

732       uptime counter - info
733          Management process uptime
734
735          Uptime in seconds of the management process
736
737       child_start counter - diag
738          Child process started
739
740          Number of times the child process has been started
741
742       child_exit counter - diag
743          Child process normal exit
744
745          Number of times the child process has been cleanly stopped
746
747       child_stop counter - diag
748          Child process unexpected exit
749
750          Number  of times the child process has exited with an unexpected re‐
751          turn code
752
753       child_died counter - diag
754          Child process died (signal)
755
756          Number of times the child process has died due to signals
757
758       child_dump counter - diag
759          Child process core dumped
760
761          Number of times the child process has produced core dumps
762
763       child_panic counter - diag
764          Child process panic
765
766          Number of times the management process has caught a child panic
767

MEMPOOL – MEMORY POOL COUNTERS

769       live gauge - debug
770          In use
771
772       pool gauge - debug
773          In Pool
774
775       sz_wanted gauge - debug
776          Size requested
777
778       sz_actual gauge - debug
779          Size allocated
780
781       allocs counter - debug
782          Allocations
783
784       frees counter - debug
785          Frees
786
787       recycle counter - debug
788          Recycled from pool
789
790       timeout counter - debug
791          Timed out from pool
792
793       toosmall counter - debug
794          Too small to recycle
795
796       surplus counter - debug
797          Too many for pool
798
799       randry counter - debug
800          Pool ran dry
801

SMA – MALLOC STEVEDORE COUNTERS

803       c_req counter - info
804          Allocator requests
805
806          Number of times the storage has been asked to provide a storage seg‐
807          ment.
808
809       c_fail counter - info
810          Allocator failures
811
812          Number of times the storage has failed to provide a storage segment.
813
814       c_bytes counter - info
815          Bytes allocated
816
817          Number of total bytes allocated by this storage.
818
819       c_freed counter - info
820          Bytes freed
821
822          Number of total bytes returned to this storage.
823
824       g_alloc gauge - info
825          Allocations outstanding
826
827          Number of storage allocations outstanding.
828
829       g_bytes gauge - info
830          Bytes outstanding
831
832          Number of bytes allocated from the storage.
833
834       g_space gauge - info
835          Bytes available
836
837          Number of bytes left in the storage.
838

SMU – UMEM STEVEDORE COUNTERS

840       c_req counter - info
841          Allocator requests
842
843          Number of times the storage has been asked to provide a storage seg‐
844          ment.
845
846       c_fail counter - info
847          Allocator failures
848
849          Number of times the storage has failed to provide a storage segment.
850
851       c_bytes counter - info
852          Bytes allocated
853
854          Number of total bytes allocated by this storage.
855
856       c_freed counter - info
857          Bytes freed
858
859          Number of total bytes returned to this storage.
860
861       g_alloc gauge - info
862          Allocations outstanding
863
864          Number of storage allocations outstanding.
865
866       g_bytes gauge - info
867          Bytes outstanding
868
869          Number of bytes allocated from the storage.
870
871       g_space gauge - info
872          Bytes available
873
874          Number of bytes left in the storage.
875

SMF – FILE STEVEDORE COUNTERS

877       c_req counter - info
878          Allocator requests
879
880          Number of times the storage has been asked to provide a storage seg‐
881          ment.
882
883       c_fail counter - info
884          Allocator failures
885
886          Number of times the storage has failed to provide a storage segment.
887
888       c_bytes counter - info
889          Bytes allocated
890
891          Number of total bytes allocated by this storage.
892
893       c_freed counter - info
894          Bytes freed
895
896          Number of total bytes returned to this storage.
897
898       g_alloc gauge - info
899          Allocations outstanding
900
901          Number of storage allocations outstanding.
902
903       g_bytes gauge - info
904          Bytes outstanding
905
906          Number of bytes allocated from the storage.
907
908       g_space gauge - info
909          Bytes available
910
911          Number of bytes left in the storage.
912
913       g_smf gauge - info
914          N struct smf
915
916       g_smf_frag gauge - info
917          N small free smf
918
919       g_smf_large gauge - info
920          N large free smf
921

VBE – BACKEND COUNTERS

923       happy bitmap - info
924          Happy health probes
925
926          Represents the last probe results as a bitmap. Happy probes are bits
927          set to 1, and the unhappy ones are set to 0. The highest bits repre‐
928          sent the oldest probes.
929
930       bereq_hdrbytes counter - info
931          Request header bytes
932
933          Total backend request header bytes sent
934
935       bereq_bodybytes counter - info
936          Request body bytes
937
938          Total backend request body bytes sent
939
940       beresp_hdrbytes counter - info
941          Response header bytes
942
943          Total backend response header bytes received
944
945       beresp_bodybytes counter - info
946          Response body bytes
947
948          Total backend response body bytes received
949
950       pipe_hdrbytes counter - info
951          Pipe request header bytes
952
953          Total request bytes sent for piped sessions
954
955       pipe_out counter - info
956          Piped bytes to backend
957
958          Total number of bytes forwarded to backend in pipe sessions
959
960       pipe_in counter - info
961          Piped bytes from backend
962
963          Total number of bytes forwarded from backend in pipe sessions
964
965       conn gauge - info
966          Concurrent connections used
967
968          The number of currently used connections to the backend. This number
969          is always less or equal to the number of connections to the  backend
970          (as,  for  example  shown as ESTABLISHED for TCP connections in net‐
971          stat) due to connection pooling.
972
973       req counter - info
974          Backend requests sent
975
976       unhealthy counter - info
977          Fetches not attempted due to backend being unhealthy
978
979       busy counter - info
980          Fetches not attempted due to backend being busy
981
982          Number of times the max_connections limit was reached
983
984       fail counter - info
985          Connections failed
986
987          Counter of failed opens. Detailed reasons are given  in  the  fail_*
988          counters (DIAG level) and in the log under the FetchError tag.
989
990          This counter is the sum of all detailed fail_* counters.
991
992          All fail_* counters may be slightly inaccurate for efficiency.
993
994       fail_eacces counter - diag
995          Connections failed with EACCES or EPERM
996
997       fail_eaddrnotavail counter - diag
998          Connections failed with EADDRNOTAVAIL
999
1000       fail_econnrefused counter - diag
1001          Connections failed with ECONNREFUSED
1002
1003       fail_enetunreach counter - diag
1004          Connections failed with ENETUNREACH
1005
1006       fail_etimedout counter - diag
1007          Connections failed ETIMEDOUT
1008
1009       fail_other counter - diag
1010          Connections failed for other reason
1011
1012       helddown counter - diag
1013          Connection opens not attempted
1014
1015          Connections not attempted during the backend_local_error_holddown or
1016          backend_remote_error_holddown interval after a  fundamental  connec‐
1017          tion issue.
1018

LCK – LOCK COUNTERS

1020          Counters  which  track  the activity in the different classes of mu‐
1021          tex-locks.
1022
1023          The counts may be slightly wrong if there are more than one lock in‐
1024          stantiated in each class (ie: .creat > 1)
1025
1026       creat counter - debug
1027          Created locks
1028
1029       destroy counter - debug
1030          Destroyed locks
1031
1032       locks counter - debug
1033          Lock Operations
1034
1035       dbg_busy counter - debug
1036          Contended lock operations
1037
1038          If the lck debug bit is set: Lock operations which returned EBUSY on
1039          the first locking attempt.
1040
1041          If the lck debug bit is unset, this counter  will  never  be  incre‐
1042          mented even if lock operations are contended.
1043
1044       dbg_try_fail counter - debug
1045          Contended trylock operations
1046
1047          If  the  lck  debug  bit  is  set: Trylock operations which returned
1048          EBUSY.
1049
1050          If the lck debug bit is unset, this counter  will  never  be  incre‐
1051          mented even if lock operations are contended.
1052

AUTHORS

1054       This  man  page  was  written  by  Lasse Karstensen, using content from
1055       vsc2rst written by Tollef Fog Heen.
1056
1057
1058
1059
1060                                                           VARNISH-COUNTERS(7)
Impressum