+ Wj(Rt^RIHt^RIt^RIt^RIt^RIHt^RIH t H t H t ]P!] 4t^xR^yR^zR^R/tR t]R ]^,R /tR R ltRRltRRltRRltR#)aWho holds the SQLite write lock on state.db right now (Linux ``/proc/locks``). The "database is locked for over Ns" failure names the victim but not the holder, and the open-descriptor scan (``hermes_state_holders``) cannot tell a reader from the one writer: every Hermes process has the DB open. SQLite's unix VFS takes ``fcntl`` byte-range locks whose offsets encode the lock kind, and the kernel exports them with the owning pid, so the holder can be named at the moment the deadline passes. ) annotationsN)Path)DictListTuplez WAL writezWAL checkpointz WAL recoverzshm dead-man switchi@PENDINGRESERVEDc(V^8dQhRRRRRRRR/#)sidecarstrstartintendreturn)formats"KC:\Users\johnn\AppData\Local\hermes\hermes-agent\hermes_state_lockowners.py __annotate__r#s( % %S % %3 %3 %cVR8XdKV\9dW!8Xd\V,#^{Tu;8:d^8:dMMW!8Xd RV^{, 2#RV RV 2#W8XdV\9d\V,#V\^,8XdR#RV RV 2#)-shmzWAL read slot z shm bytes -z SHARED rangez db bytes )_SHM_LOCK_NAMES_MAIN_LOCK_NAMES _PENDING_BYTE)r r rs&&&r_describe_ranger#s& O # "5) ) % 3 3<#ECK=1 1E7!C5)) |!11&& !! ugQse $$rc$V^8dQhRRRRRR/#)r textr inodeszdict[tuple[int, int], str]rzlist[tuple[int, str, str]]r)rs"rrr1s#3(BGarc F.pVP4FpVP4p\V4^8gV^,R8XdK3VR,wrVrxrrVR8wdKJV PR4wrp\P!\ V ^4\ V^44pV\ V43p\ V 4p\ V 4pV R8XdTM \ V 4pTPT4pTfKTPT\TTT4T34K V# \ dEK i;i)a/``(pid, lock kind, sidecar)`` for every WRITE lock on one of ``inodes``. ``inodes`` maps ``(st_dev, st_ino)`` to ``""`` (main file), ``"-wal"`` or ``"-shm"``. Read locks are dropped: they never block a writer in WAL mode. An OFD lock is reported with pid ``-1`` (the kernel does not export its owner). z->:NNWRITE:EOF) splitlinessplitlenosmakedevr ValueErrorgetappendr)rrfoundlinefields__kind_clsaccesspid_sidentstart_send_smajor_sminor_sino_sdevkeypidr rr s&& rparse_proc_locksr>1s)+E! v;?fQi4/ ?Ebz<$g W   &+kk#&6 #Ge**S"-s7B/?@CE #Ce*CLE E>%s5zC**S/ ?  c?7E3?IJ)"* L    sA6D D D cV^8dQhRR/#)r rz list[str]r)rs"rrrQsIrc 6a \PPR4'g.#\PP \P !V44p/pR F:p\P!W,4pY2TPTP3&K< .o \^4FXp\RRRR7;_uu_4pVP4pRRR4S PT 3Rl\XT444KZ ^RIHp.p S FIwrpV ^8d V!V 4MR p V P%V R V R \'W,4P( 24KK V # \dEKi;i +'giL;i \d.uu#i;i) zOperator-facing lines naming the processes that hold a write-class lock on ``db_path``. Empty when nothing is held or the platform has no ``/proc/locks``. linuxz /proc/locksasciireplace)encodingerrorsNc38<"TFqS9gK VxK R#5i)Nr).0entryhelds& r .state_db_write_lock_holders..is['EeVZIZEE'Es )describe_holder_pidz.OFD lock, owner pid not exported by the kernelz holds z lock on )z-walr)sysplatform startswithr(pathrealpathfspathstatOSErrorst_devst_inorangeopenreadextendr>hermes_state_holdersrLr,rname)db_pathbaserr str0handlerrLlinesr=kindwhorIs& @rstate_db_write_lock_holdersreQs_ << " "7 + + 77  BIIg. /D)+F' (B*1 299%& ((*D 1X mgiHHF{{}I ['7f'E[[ 9 E"7*-'!#&7g uGD643G3L3L2MNO# L'   IH I sB$E /FE3F E0/E03 F >FF FFc V^8dQhRRRR/#)r patience_sfloatrNoner)rs"rrrss   $ rc\V4pV'dRPV4pMRp\PRV\ V4P V4R#)zQOne WARNING naming the write-lock holders at the moment a writer gave up waiting.z | zMno write-class lock held at the deadline (the holder released just before it)z2state.db write lock unavailable for %.0fs (%s): %sN)rejoinloggerwarningrr])r^rgholdersdetails&& rlog_write_lock_holdersrpssB)'2GG$` NNrerprrrrxs# $$   8 $    9Az %@D r