ci-server-engineer/knowledge/manuals/IM36K10L20-01EN.md

manual	title	edition	version	status
IM 36K10L20-01EN	History Management Engineering	4th Edition, Feb. 2026 (R1.05; first summarized from the 3rd Edition, Sep. 2023, R1.03)	R1.05 (first written for R1.03)	R1.03 - hisdbg read-only options, rollover and history file layout verified on the test system 2026-10-01; archive (hasman/hasexe/hasdbg), hisrecover, hisrep, hiswrn from docs; setup files checked read-only. R1.05 - text diff against R1.03 shows no content change (2026-10-08); has.sup, hisini.sup and the HISTORY program names compared with snapshots of the R1.05 test server (same, apart from instance/licence lines); history_start.cmd / history_stop.cmd differ and hisdbg was not run on R1.05 (awaiting user verification, ISS-100/101)

History Management Engineering (R1.05)

R1.05: the 4th Edition (Feb. 2026) has no content change against the R1.03 edition (full text diff, 2026-10-08; only the edition date and contact page). Section numbers are unchanged. Installed files on the R1.05 test server (compared 2026-10-08 with snapshots taken that day): same program file names (his, hisdbg, hisrep, hisrecover, hiswrn, hasman, hasexe, hasdbg, lghumh …), same has.sup, hisini.sup differs only in instance and licence lines; history_start.cmd and history_stop.cmd differ (content not yet read, ISS-100).

HISTORY Service schedules and administers history groups (storage groups) and their units (files) for all services that keep history: item history (ITEM Service), the chronological alarm overview (ALH), the audit trail, the system log and report definitions. It does not know the contents of the files; the client services write and read them. Section numbers refer to the manual. Engineering Module forms and item storage group types: IM36K10C40-01EN.md; item history processes (ITH*): IM36K10L35-01EN.md.

1. Processes and data (§1)

Process	Role
HIS	History manager: schedules start / stop / rollover / delete commands, keeps the administration, serves the message interface. No setup file, no arguments
HASMAN	Archive manager (automatic daily archive, manual archive, retrieve); has.sup
HASEXE	Archive executor: accesses the archive device; has.sup
LGHUMH	Copies message log entries into the SYS_LOG group (started by history_start.cmd on the host node)

Installed tls\com\history_start.cmd (R1.03; the R1.05 file differs, ISS-100): creates %TLS_HIS_PATH%\has, runs hisrep automatically at every start when his_units.dat exists, then starts his, lghumh, hasexe, hasman.

Administration (DATABASE Service files in tls\dat): his_proc.dat (every client that ever logged in; they get history manager up at start), his_groups.dat (group definitions and schedule), his_units.dat (all units with start, stop, delete and warning times).

Archive status item (has.sup STATUS_ITEM): statuses ARCHIVE_A, ARCHIVE_M, ARCHIVE_R, MOUNT_TAPE, INIT_TAPE, REG_TAPE, SKIP_TAPE, CANCEL, READY (value 1, normal) and FAILED (value 0, Alarm 1). A storage group can be excluded from archiving (HIS_GROUP_DF EXCLUDE_ARCHIVE).

2. History files (§2.1)

• Group number = node + sequence; unit number = unit sequence (incremented when the group is modified) + data sequence (incremented at each rollover).
• File names use base-32 digits (0–9, a–v): h<nn><ggg>000000.his group description, h<nn><ggg><uuu>000.his unit description, h<nn><ggg><uuu><ddd>.his unit data; reports also …000000.rpt. Each is a DATABASE Service file with .d01 / .k01 / .k02 companions.
• Verified (%TLS_ALLUSERS_PATH%\tls\his\, hisdbg): CIENG_EVENT = group 1,54 → h0101m…; CIENG_SCAN10S 1,55 → h0101n…; CIENG_AGG 1,56 → h0101o… plus extra h0101ozzz… files for the aggregates. After a CI Server restart the unit sequence jumped from 1 to 3 (consistent with hisrep at start, which increases unit sequences by 1).

3. Debugger hisdbg (§3.1)

Starts by asking for the node (blank = local). Menu (installed: -1 … 22):

No.	Action	Use
1	Statistics (groups, logged-in clients, commands, unit starts/stops/rollovers/deletions)	read-only
2 / 21 / 22	Group info (last group / by name / by number): path, group number, next unit and file, his_type, opt_type, start, stop, first rollover, interval, life time, warn time, GMT/LCT rollover, archiving flag, lock state	read-only
3	Unit info of the current group: unit, data, file, start, stop, delete (GMT)	read-only
6 / 7	Logged-in clients / command interests of one client	read-only
11	Show unknown history files	read-only
20	All groups: name, number, type	read-only
4 / 8 / 10 / 12	Delete unit / delete group (no active unit) / restore unit (file must exist) / delete unknown files	changes history — never without approval
5	Force rollover (same as Engineering Module Force rollover; not within 2 min of the last one)	changes files
9	Change schedule (start, stop, next rollover, interval, life time, GMT rollover, warn time)	changes the group

History types: 21 alarm (ALH), 24 item, 6 audit (ADT), 1058 report definitions; on the test system also 4 (STATION_STAT) and 36 (SYS_LOG), not listed in the manual (ISS-039). opt_type shows the item collection/storage type (the manual's table is garbled in the PDF); verified: Scan/Time 1 1 (CIENG_SCAN10S), Event/Item 2 2 (CIENG_EVENT), Aggregation 4 4 (CIENG_AGG). Times are shown in GMT.

Verified on the R1.03 test system (2026-10-01): 45 groups, 30 logged-in clients (ITH*, ALH, ADTSTO, DSSSDA/DSSEVT, LGHUMH, EQPIOSTAT, RPTMAN, HIS); group info of CIENG_EVENT (path C:\Users\Public\Yokogawa\tls\his\, rollover daily at 00:00 GMT, life 7 days, GMT rollover, included in archiving); force rollover of CIENG_EVENT closed unit 3,1 and started 3,2.

4. Other utilities (§3.2–§3.5)

Utility	Use
hisrecover <file> <start> <stop>	Make an unknown unit data file known again (start/stop LCT dd-mmm-yy hh:mm:ss); group and unit description must be known
hiswrn	Reset deletion warning times of units saved with a backup prefix
hisrep	Repair unit administration after crashes or full disks: increase unit sequences, close units with stop time 0 using the last time-control record. Runs at every CI Server start
hasdbg	Archive debugger: 1 initialise, 2 mount, 3 register medium, 4/5 get/set automatic archive time, 6 manual archive, 7 retrieve, 8 overview, 9 cancel, 10 delete a medium from the administration, 11 list media, 20 dump medium

5. Archive (§2.2)

• HASMAN decides what to archive (daily automatic archive or on request), HASEXE writes it. Devices: tape (TAPE-DRIVE, Windows default //./TAPE0) or a disk path (CD-DRIVE: local disk, network share, USB, UDF-formatted CD/DVD). One device at a time; restart after changing has.sup. Archive size ≈ 20 % of the history data.
• Administration: has_idx_act.dat (initialise / archive actions) and has_idx_grp.dat (groups and units per medium), also copied to each medium. If corrupt: stop CI Server, delete both from tls\dat, start, register the latest initialised medium (and the latest used one). Remove damaged media with hasdbg 10.
• Installed has.sup (same on R1.03 and R1.05): no drive configured (archiving not set up), COMPRESSION-RATIO 6, SH_EOT / SH_NO_TAPE shell signals. Other keywords: TAPE-SIZE, STATUS_ITEM, HISTORY-PATH, TEMPORARY-PATH (scratch folder for big archives), HIS, SH_WRITE_PROT, SH_IGNORE, SH_KILL, DEBUG.

Not practiced: archiving and retrieve (no archive device; would write to disk and change the archive administration), hisrecover, hiswrn, running hisrep by hand.

R1.05 start scripts (V6, 2026-10-09)

The R1.05 history_start.cmd / item_start.cmd (and stop scripts) only add the extension-script call ?*_<script>.cmd (ISS-121); the processes started are the same as on R1.03 (ISS-100).