| 8 years agoadrian_waters commented on review 21010 for perforce-software-sdp:dev@tom_tyler - hi Tom - will you get a chance to review this over next couple of days so I can tidy up by end of week? | |
|
|
| 8 years agoadrian_waters requested review 21010 for perforce-software-sdp:devModified 'cmd' function to exclude 'error' lines just caused by no files, or more likely no shelves, in a depot - these are not errors but cause confu...sion when reviewing the p4verify.log file
- tom, see you have the file checked out so not sure whether you've still got further changes to make - the modified cmd function could do with tidying up a bit « | |
|
|
| 8 years agoadrian_waters committed change 21007 into perforce-software-sdp:dev Fix specification of depot path in get_verify_rev_range when -recent used | |
|
|
| 8 years agoadrian_waters created job000635 forp4verify.sh records 'error' lines into its logfile where depots do not contain shelved changes; this is not an error, but is confusing to client...s; it would be better to filter out such lines so they are not recorded in the log. « | |
|
|
| 8 years agoadrian_waters requested review 20992 for perforce-software-sdp:devRefactor around changes to support -recent arg - had broken code even where -recent not used because of unbounded variable ($d); use of the -rec...ent option also fails as 'verify -S' not valid for shelved files (only @= is supported) ; stripping out archive,unload+remote depots also not working + get_verify_rev_range used depot name incorrectly in changes (eg, changed -m 1 depot/...
rather than //depot/...)
Raised job000634 for changes.
Further aim is to modify script for depots with no shelved files does not result in an 'error' line being written to p4verify.log - it confuses clients - will do that as a separate change. « | |
|
|
| 8 years agoadrian_waters modified job000634 forp4verify.sh is currently broken - use of an unbounded variable ($d); use of the -recent option fails as for verify -S you cannot specify a range... (only @= is supported) ; stripping out archive,unload+remote depots also not working.
get_verofy_rev_range used depot name incorrectly in changes « | |
|
|
| 8 years agoadrian_waters created job000634 forp4verify.sh is currently broken - use of an unbounded variable ($d); use of the -recent option fails as for verify -S you cannot specify a range... (only @= is supported) ; stripping out archive,unload+remote depots also not working. « | |
|
|
| 8 years agoadrian_waters committed change 20964 into perforce-software-sdp:dev Include removal of the p4login.*.log files in daily cleanup | |
|
|
| 8 years agoadrian_waters modified job000626 forp4login.*.log files are not being cleaned up by the standard clean-up mechanism (remove_old_logs in backup_function.sh).
Note that as there may be mu...ltiple p4login log files per day keeping the last 'KEEPLOGS' is not a good solution - better to keep that last 'x' days « | |
|
|
| 8 years agoadrian_waters modified job000626 forp4login.*.log files are not being cleaned up by the standard clean-up mechanism (remove_old_logs in backup_function.sh).
Note that as there may be mu...ltiple p4login log files per day keeping the last 'KEEPLOGS' is not a good solution - better to keep that last 'x' days « | |
|
|
| 8 years agoadrian_waters created job000626 forp4login.*.log files are not being cleaned up by the standard clean-up mechanism (remove_old_logs in backup_function.sh).
Note that as there may be mu...ltiple p4login log files per day keeping the last 'KEEPLOGS' is not a good solution - better to keep that last 'x' days « | |
|
|
| 8 years agoadrian_waters approved review 20847 for perforce-software-sdp:dev Minor doc update (copyright string date);
Rewiew requests to testing purposes - please leave. | |
|
|
| 8 years agoadrian_waters created job000623 forFor a replica, if P4TARGET included ssl:, instance_vars.template issue parsing the 'target hostname' from the P4TARGET string resulted in invalid sett...ings being generated in p4_N.vars - eg: P4MASTER=ssl « | |
|
|
| 8 years agoadrian_waters requested review 20918 for perforce-software-sdp:mainIssue setting P4MASTER when using ssl
Code was not catering for scenarios where the P4TARGET setting (extracted from the db) included the ssl: prefix..., as it stripped everything from the first : onwards (leaving P4MASTER as 'ssl:'); now does this in two parts - firstly to strip off from the last :, and then to strip anything up to the first : (if there is one).
Code in this area was re-worked in previous change, but problem already existed previous releases. « | |
|
|
| 8 years agoadrian_waters committed change 20849 into perforce-software-sdp:dev Minor doc update (copyright string date); | |
|
|
| 8 years agoadrian_waters requested review 20847 for perforce-software-sdp:dev Minor doc update (copyright string date);
Rewiew requests to testing purposes - please leave. | |
|
|
| 9 years agoadrian_waters requested review 19305 for perforce-software-sdp:devIf creating a replica that shares depot files with master, don't do chown/chmod on the depot files as this can take a significant time when migrating... existing Helix servers into the SDP; In addition, issue 'warning' message that chown/chmod could take some time to complete so user is aware of processing being carried out « | |
|
|
| 9 years agoadrian_waters committed change 19176 into perforce_software-p4training:main Update for 2016.1 | |
|
|
| 9 years agoadrian_waters requested review 19062 for perforce-software-sdp:dev | |
|
|
| 9 years agoadrian_waters commented on review 18563 for perforce-software-sdp:devAny implications if upgrading an existing SDP install? | |
|
|
| 9 years agoadrian_waters requested review 18343 for Trap/default $CUT / $GREP if not set | |
|
|
| 9 years agoadrian_waters created job000457 forAdd a 'Step Mode' option (-S arg) to prompt to continue after each action - useful for demonstrating / monitoring the resultant actions in server of e...ach config item « | |
|
|
| 9 years agoadrian_waters requested review 18338 forAdd 'Step Mode' (-S argument) so process pauses (for user input) between each configuration file action
#review @ttyler | |
|
|
| 9 years agoadrian_waters commented on review 18275 for perforce-software-sdp:devLooks as though you overwrite the log each time (echo "$numfiles log files found" > $HTDLOG) - was that your intention - wouldn't it be better to d ...Looks as though you overwrite the log each time (echo "$numfiles log files found" > $HTDLOG) - was that your intention - wouldn't it be better to date-stamp the entries into the logfile & maintain the log on an ongoing basis? « | |
|
|
| 9 years agoadrian_waters approved review 16795 for perforce-software-sdp:dev Added the audit log as a default structured log and set the rotation to keep 31 days. | |
|
|
| 9 years agoadrian_waters commented on review 16795 for perforce-software-sdp:devchanges look ok, noting that in the .sh version we're testing the p4d version to check if (one of) the new configurables is appropriate, whereas we do ...changes look ok, noting that in the .sh version we're testing the p4d version to check if (one of) the new configurables is appropriate, whereas we don't in the .bat so could try setting for a pre 2015.2 p4d « | |
|
|
| 9 years agoadrian_waters committed change 16689 into perforce-software-sdp:main Include information for using systemd/systemctl for enabling services | |
|
|
| 9 years agoadrian_waters approved review 16557 for perforce-software-sdp:mainChange commented out entry P4PORT->P4_PORT so if uncomment the block of env settings if want auto-generated port numbers it doesn't fail unbound te...st (later code now used P4_PORT) « | |
|
|
| 9 years agoadrian_waters requested review 16557 for perforce-software-sdp:mainChange commented out entry P4PORT->P4_PORT so if uncomment the block of env settings if want auto-generated port numbers it doesn't fail unbound te...st (later code now used P4_PORT) « | |
|
|
| 9 years agoadrian_waters requested review 16556 for perforce-software-sdp:main Added lbr.autocompress=1 and changed limit daily checkpoints to false. | |
|
|
| 9 years agoadrian_waters left review 16527 for perforce-software-sdp:mainChanged storage for /p4/<n>/tmp dir to the /logs volume.
The /logs volume is typically local, and is not backed up. Both
of these charac...teristics are desirable for TMP/P4TMP. We don't
want to pay the typically high latency tax of shared storage solutions
used for the /depotdata for TMP/P4TMP. And we don't want to
needlessly backup up a folder whose name implies it's not backed
up.
To Do:
* Make the corresponding change in the Windows SDP.
* Adjust docs as needed.
#review @russell_jackson « | |
|
|
| 9 years agoadrian_waters joined review 16527 for perforce-software-sdp:mainChanged storage for /p4/<n>/tmp dir to the /logs volume.
The /logs volume is typically local, and is not backed up. Both
of these charac...teristics are desirable for TMP/P4TMP. We don't
want to pay the typically high latency tax of shared storage solutions
used for the /depotdata for TMP/P4TMP. And we don't want to
needlessly backup up a folder whose name implies it's not backed
up.
To Do:
* Make the corresponding change in the Windows SDP.
* Adjust docs as needed.
#review @russell_jackson « | |
|
|
| 10 years agoadrian_waters committed change 15376 into perforce-software-sdp:dev formatting only - fix spacing; there's inconsistent use of tabs/spaces throughout the file - needs cleanup at some point. | |
|
|
| 10 years agoadrian_waters committed change 15375 into perforce-software-sdp:dev Routine merge-down from main->dev | |
|
|
| 10 years agoadrian_waters committed change 15374 into perforce-software-sdp:dev- Ensure backup scripts are run as the OSUSER (to prevent accidental running as root);
- in scripts where LOGFILE value is changed from the 'ch...eckpoint.log' set by set_vars, ensure the new assignment is before check_dirs is called, otherwise errors could be written to the 'wrong' log
- in 'die()' - detect if running from terminal & also send output to stderr « | |
|
|
| 10 years agoadrian_waters updated description of review 15364 for perforce-software-sdp:dev- Ensure backup scripts are run as the OSUSER (to prevent accidental running as root);
- in scripts where LOGFILE value is changed from the 'ch...eckpoint.log' set by set_vars, ensure the new assignment is before check_dirs is called, otherwise errors could be written to the 'wrong' log
- in 'die()' - detect if running from terminal & also send output to stderr « | |
|
|
| 10 years agoadrian_waters updated files in review 15364 for perforce-software-sdp:dev- Ensure backup scripts are run as the OSUSER (to prevent accidental running as root);
- in scripts where LOGFILE value is changed from the 'ch...eckpoint.log' set by set_vars, ensure the new assignment is before check_dirs is called, otherwise errors could be written to the 'wrong' log « | |
|
|
| 10 years agoadrian_waters requested review 15364 for perforce-software-sdp:dev- Ensure backup scripts are run as the OSUSER (to prevent accidental running as root);
- in scripts where LOGFILE value is changed from the 'ch...eckpoint.log' set by set_vars, ensure the new assignment is before check_dirs is called, otherwise errors could be written to the 'wrong' log « | |
|
|
| 10 years agoadrian_waters committed change 15290 into perforce-software-sdp:mainRemove ref to incremental verify, mention internal LDAP integration, clarify OSUSER/OSGROUP settings, formatting tidyups
- p4verify_incremental.sh... was deprecated in chgl 11466 so remove from doc
- in initial server setup section, referred to user/group p4admin (& ADMINUSER); change to perforce/perforce to reflect mkdirs.sh
- mention p4 ldap option for 2014.2+ « | |
|
|
| 10 years agoadrian_waters created job000235 for perforce-software-p4convert:backdoor cvs conversion generates multiple db.domain records for each label - not a major problem as they get squashed to 1 when generated journal loa...ded « | |
|
|
| 10 years agoadrian_waters created job000234 for perforce-software-p4convert:Add validation to user map file to trap spaces in the destination name; backdoor convert+journal load will allow this through, but a front-door... import will fail « | |
|
|
| 10 years agoadrian_waters committed change 11909 into perforce-software-sdp:mainChange commented out entry P4PORT->P4_PORT so if uncomment the block of env settings if want auto-generated port numbers it doesn't fail unbound te...st (later code now used P4_PORT) « | |
|
|
| 10 years agoadrian_waters committed change 11908 into perforce-software-sdp:main Use set -u to trap unbounded variables | |
|
|
| 10 years agoadrian_waters committed change 11906 into perforce-software-sdp:mainfix @11903 - to ensure comparing /p4 against /tmp/p4 when in test mode | |
|
|
| 10 years agoadrian_waters committed change 11905 into perforce-software-sdp:main Fix creation of the reset_sdp.sh script by the bootstrap script - it was copying it to /depotdata/sdp/sdp | |
|
|
| 10 years agoadrian_waters committed change 11903 into perforce-software-sdp:mainFix issue introduced in @11877 - the echo'd diff statements at end should compare /p4 against $P4DIR when in test mode | |
|
|
| 10 years agoadrian_waters committed change 11885 into perforce-software-sdp:mainAdd setting to trap unbounded variables; caught/fixed one instance of error around test mode, so enhanced 'usage' check & test mode processi...ng; excluded setting ownership/perms on /tmp/p4/sdp in test mode « | |
|
|
| 10 years agoadrian_waters committed change 11884 into perforce-software-sdp:main fix comparison to support non-numeric instance | |
|
|
| 10 years agoadrian_waters committed change 11883 into perforce-software-sdp:main fix unbounded use of P4BROKER_PORT_END (variable removed in #2) | |
|
|
| 10 years agoadrian_waters committed change 11724 into perforce-software-sdp:main SDP/mkdirs.sh - fix bug that prevents MAILTO being subsituted in the crontab / crontab.replica files | |
|
|
