| 7 months agosuper-tom_tyler updated project (AutoMerge) | |
|
|
| 7 months agosuper-tom_tyler updated project (AutoMerge) | |
|
|
| 4 years agosuper-tom_tyler modified SDP-574 fordaily_checkpoint.sh hangs if journalPrefix is wrong.
Detail: The truncate_journal() in backup_functions.sh hangs if
journalPrefix is misconfigured.
...
There is a logic block in truncate_journal that is intended to wait
for a 'p4 -p $P4MASTERPORT admin journal' to fully complete processing
before the function returns. It does so by waiting for the numbered
journal file to appear in the checkpoints folder. This works as
intended when the journalPrefix is correct.
However, if the journalPrefix is wrong, the wait loop still occurs,
causing a hang waiting forever for a journal file that is never going
to appear. « | |
|
|
| 6 years agosuper-tom_tyler commented on review 25741 (monitor_metrics.sh, line 49) for perforce-software-sdp:devPerhaps compare against template.sh line, which uses:
msg "That took $((SECONDS/3600)) hours $((SECONDS%3600/60)) minutes $((SECONDS%60)) seconds.\n" ...Perhaps compare against template.sh line, which uses:
msg "That took $((SECONDS/3600)) hours $((SECONDS%3600/60)) minutes $((SECONDS%60)) seconds.\n"
If we need base10, we should tweak that in template.sh and other scripts too. Or maybe we only need to ensure base10 if not using the built-in SECONDS variable? « | |
|
|
| 6 years agosuper-tom_tyler created P4XFER-1 for perforce-software-p4transfer:Add an interactive wizard of some kind to greatly simplify
setup of a P4Transfer configuration.
It should query the user for a working directory (t...he root
of the workspaces and home of the transfer.cfg file), and
then query them for details of the source and target servers.
It should then ask questions that determine the set of default
flags to be used for that configuration (e.g. -k, --ignore,
etc.), and somehow persist that information, e.g. perhaps
generating a 'P4TransferWrapper.py' wrapper script that
just calls P4Transfer.py withose flags.
Then the wizard should generate the workspaces to be used
with P4Transfer on both the source and target servers. It
should login in using the credentials provided (relying on a
ticket if no password is specified, just as with normal
P4Transfer.py operation).
The workspaces should be configured with best practices for a
workspace used P4Transfer. By "best practices," it should,
for example, determine whether 'noallwrite' or 'allwrite'
option and/or the submitunchanged/leaveunchanged option is best
for a workspace to be used with P4Transfer. If there is no
best practice, just pick one. « | |
|
|
| 6 years agosuper-tom_tyler created P4XFER-2 for perforce-software-p4transfer:Add an interactive wizard of some kind to greatly simplify
setup of a P4Transfer configuration.
It should query the user for a working directory (t...he root
of the workspaces and home of the transfer.cfg file), and
then query them for details of the source and target servers.
It should then ask questions that determine the set of default
flags to be used for that configuration (e.g. -k, --ignore,
etc.), and somehow persist that information, e.g. perhaps
generating a 'P4TransferWrapper.py' wrapper script that
just calls P4Transfer.py withose flags.
Then the wizard should generate the workspaces to be used
with P4Transfer on both the source and target servers. It
should login in using the credentials provided (relying on a
ticket if no password is specified, just as with normal
P4Transfer.py operation).
The workspaces should be configured with best practices for a
workspace used P4Transfer. By "best practices," it should,
for example, determine whether 'noallwrite' or 'allwrite'
option and/or the submitunchanged/leaveunchanged option is best
for a workspace to be used with P4Transfer. If there is no
best practice, just pick one. « | |
|
|
| 7 years agosuper-tom_tyler committed change 23498 into Added license. | |
|
|
| 7 years agosuper-tom_tyler modified SDP-58 forTake full advantage of systemd features.
With job SDP-13, basic systemd support was added to the SDP,
enabling SDP to play nice on Linux OS veresion...s where systemd
replaces SYS5 init scirpts. However, the goal then was merely
to add basic support.
This job captures the hope to explore and take full advantage of
systemd features to optimum benefit of the SDP. Exatly what
should be done requires getting fully up to speed on systemd
strengths, features, limitations, etc.
See also: SDP-241 « | |
|
|
| 7 years agosuper-tom_tyler modified SDP-258 for Doing 'source p4_vars 1' hangs if a live checkpoint is running. | |
|
|
| 8 years agosuper-tom_tyler modified CBD-23 forImplement flexbile feature toggles.
Flexible means feature can be enabled globally, per-component,
or in some cases more granular, e.g. per-stream.
...
Example: Make the versioning of the Remapped field
(CBD-17) be configurable as always on, always off,
or always on but limited to the main branch. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-24 forEnable bi-directional stream-spec/*.cbdsst workflow.
Provide a workflow that allows developers to modfiy CBD configurations
via traditional stream s...pec edits, or via changes to the *.cbdsst file.
A change in either updates the other implicitly via trigger actions.
This change will help make CBD automation somewhat more transparent.
Implementation Detail: Presently, the CBD trims version specifiers
from the versioned form of the stream spec (the *.cbdsst file). This
is done for a few reasons:
* Earlier versions of P4D didn't support version specifiers in
stream specs, so they had to be trimmed off or the server would
reject them. P4D 2015.1 supports changelists only as revision
specifiers.
* CBD supports any valid revision specifier, while P4D 2015.1
supports only changelists. P4D 2015.2 may add support for
automatic labels and timestamps. That leaves only static labels
(important) and workspaces (not so import) as revision
specifiers supported by these CBD scripts outside of P4D.
Decision:
* CBD automation will detect P4D version:
o If P4D < 2014.1, strip all revision specifiers going from
*.cbdsst -> live stream spec.
o If P4D < 2014.2, strip all revision specifiers going from
*.cbdsst -> live stream spec except changelist numbers.
o If P4D < 2015.2, strip all revision specifiers going from
*.cbdsst -> live stream spec except changelist numbers and
automatic labels.
Warning: The ability to update stream specs directly had proven a
helpful workaround to earlier CBD issues which have since been
eliminated. This change would eliminate that flexibility, though it
seems it is longer be needed. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-31 forChange default sync rev for '...' on a back-in-type sync.
Change the default revision used for a 'p4 sync' when a
back-in-time sync is done, i.e. a... sync with a revision
specifier, thus selecting an older version of the
*.cbdsst file (which represents the state of the versioned
stream spec at the old revision).
Currently, when a back-in-time sync is done, the sync
uses the correct specified revisions when they are
specified explicitly in the older version of the *.cbdsst
file, e.g. a Paths: field containing:
import foo/... //foo/main/...@275
However, if no revision is explicilty specified, e.g.:
import foo/... //foo/main/...
the current behvaior is simply to pass the '...' along to
the server, resulting in the head revisions being
selected for a sync to files in //foo/main/...
With this job, the desired new behvaior is to make it
so the implicit default, when no revision is specified,
becomes the point specified changelist number.
So, if a user does:
p4 sync @500
and that results in syncing a //ace/main/ace.cbdsst@500,
and in that revision there is a Paths: field containing:
import foo/... //foo/main/...
then desired behavior is to translate that sync to the
equivalent of:
p4 sync //foo/main/...@500
== OPTIONAL ==
For visual optimization, the @500 could be replaced with
the latest changelist up to @500 that actually affected
files in the path //foo/main..., and so it might
actually get translated to something like:
p4 sync //foo/main/...@275
assuming @275 is the latest changelist up to @500 that
affected files in //foo/main/...
In all cases, 'p4 sync' behavior also applies to 'flush'
and 'update' commands (which is current behavior). « | |
|
|
| 8 years agosuper-tom_tyler created CBD-37 forAlways On mode not working correctly with StreamDepth>1.
Hi Tom,
I ran in to another issue. Always On is broken now. Expected result is
//...DeepPlayground/DeepShared1/lib1/libs/...@1207340
p4 sync --parallel=0 //DeepPlayground/DeepShared1/lib1/...#head
INFO: [CBD sync ['--parallel=0'], 'Path [//DeepPlayground/DeepShared1/lib1/...] is out of CBD scope. Assuming #head.'].
//DeepPlayground/DeepShared1/lib1/libs/readme.txt updated C:\Users\mpadmanaban\Perforce\mpadmanaban_WS-2UA2220HRW_main_2801\libs\readme.txt
1 file updated
1 warning reported
INFO: [CBD sync ['--parallel=0'], 'Path [//DeepPlayground/DeepShared1/lib1/...] is out of CBD scope. Assuming #head.'].
Thanks,
- Magesh « | |
|
|
| 8 years agosuper-tom_tyler modified job000519 for Sync on stream with import fails to pick up import when StreamDepth>1. | |
|
|
| 8 years agosuper-tom_tyler modified CBD-36 for Sync on stream with import fails to pick up import when StreamDepth>1. | |
|
|
| 8 years agosuper-tom_tyler modified CBD-36 for Sync on stream with import fails to pick up import when StreamDepth>1. | |
|
|
| 8 years agosuper-tom_tyler created CBD-35 forSync above stream root with StreamDepth>1 gives annoying but harmless warnings.
p4 sync --parallel=0 //DeepPlayground/Components/...#head
INFO: [...CBD sync ['--parallel=0'], 'Path [//DeepPlayground/Components/...] is out of CBD scope. Assuming #head.'].
//DeepPlayground/Components/main/DeepPlayground-Components.cbdsst added C:\Users\mpadmanaban\Perforce\mpadmanaban_laptop\DeepPlayground-Components.cbdsst
1 file added
1 warning reported
INFO: [CBD sync ['--parallel=0'], 'Path [//DeepPlayground/Components/...] is out of CBD scope. Assuming #head.']. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-34 forPath collision with deep depots (StreamDepth>1).
I think we need to add an error check to make sure that cbd
key *_path0 is not modified after be...ing initialized. e.g.:
Just for test, I tried the following 2 streams and it confused the cbd
script, because there is a collision.
//DeepThought/Components/tools_test1_main
//DeepThought/Components_tools/test1_main « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-33 for Paramter to logging module should be filemode rather than mode. | |
|
|
| 8 years agosuper-tom_tyler modified CBD-33 for Paramter to logging module should be filemode rather than mode. | |
|
|
| 8 years agosuper-tom_tyler modified CBD-32 for Support StreamDepth > 1. | |
|
|
| 8 years agosuper-tom_tyler created CBD-32 for Support StreamDepth > 1. | |
|
|
| 8 years agosuper-tom_tyler created CBD-31 forChange default sync rev for '...' on a back-in-type sync.
Change the default revision used for a 'p4 sync' when a
back-in-time sync is done, i.e. a... sync with a revision
specifier, thus selecting an older version of the
*.cbdsst file (which represents the state of the versioned
stream spec at the old revision).
Currently, when a back-in-time sync is done, the sync
uses the correct specified revisions when they are
specified explicitly in the older version of the *.cbdsst
file, e.g. a Paths: field containing:
import foo/... //foo/main/...@275
However, if no revision is explicilty specified, e.g.:
import foo/... //foo/main/...
the current behvaior is simply to pass the '...' along to
the server, resulting in the head revisions being
selected for a sync to files in //foo/main/...
With this job, the desired new behvaior is to make it
so the implicit default, when no revision is specified,
becomes the point specified changelist number.
So, if a user does:
p4 sync @500
and that results in syncing a //ace/main/ace.cbdsst@500,
and in that revision there is a Paths: field containing:
import foo/... //foo/main/...
then desired behavior is to translate that sync to the
equivalent of:
p4 sync //foo/main/...@500
== OPTIONAL ==
For visual optimization, the @500 could be replaced with
the latest changelist up to @500 that actually affected
files in the path //foo/main..., and so it might
actually get translated to something like:
p4 sync //foo/main/...@275
assuming @275 is the latest changelist up to @500 that
affected files in //foo/main/...
In all cases, 'p4 sync' behavior also applies to 'flush'
and 'update' commands (which is current behavior). « | |
|
|
| 8 years agosuper-tom_tyler created CBD-30 forCBD should allow for case-insenstive path comparisons.
The case
of the server should be detected, and CBD case-insensitive path
comparisons sh...ould be used if the given server is case insensitive. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-29 forFrom Windows clients, if your drive letter is uppercase in Perforce
metadta (e.g. 'C:\MyWS'), but your current working directory has a lowercase
lett...er (e.g. c:\MyWs'), CBD gives an error when trying to sync:
ERROR: CBD was not able to translate path [pb/...] to depot syntax.
If your drive letter is lowercase in the metadata, it alwasy works.
A case mismatch for the drive letter should not matter.
The expected behavior is to simply ignore the case of the drive
letter. The comparison for the rest of the path shout remain
case sensitive. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-28 forCMD line: linux and windows) Imported lines used to be able to sync with relative folder
p4 sync componentA\...@99999999999999 should pull down the... head of componentA but it can't
figure out that the componentA is part of the import spec now. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-27 forP4V on Linux: Sync from workspace root does nothing, ignores CBD altogether.
(linux cmd is OK)
p4 sync /root/of/workspace/...#head
no files update...d
1 warning reported
INFO: ['CBD sync []', '/root/of/workspace/...#head']. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-26 forFix ticket issues: Do 'p4 login' if necessary.
Presently, the CBD automation presumes it has a valid ticket. Add a
'p4 login -s' ticket... status check, and if it fails, attempt a 'p4 login'
using credentails available via a mechanism similer to that used by
the SDP to ensure checkpoints run reliably.
This check needs to be added to BOTH the update trigger and the broker
filter script.
This change will increase the coupling betweeen CBD and the SDP, which
has been kept to a minimum thus far. That's acceptable. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-25 forSync //path/...@Label always gets latest files w/p4d 2015.1.
I noticed some discrepancy in Stream sync using Labels with workspace.
For example,&n...bsp; When I sync using the command p4 sync //<workspace>/...@<Label>,
It always gets the latest code instead of the labelled code.
When I try to sync using p4 sync //<workspace>/*@<Label>, I get the correct files at the top level.
When I try to sync using p4 sync //<depot_path>/...@<Label>, I get the correct files.
Command used:
p4 sync //jam/...@LABEL <-always gets the latest files
p4 sync //jam/*@LABEL <- works
p4 sync //jam//main/...@LABEL <- works « | |
|
|
| 8 years agosuper-tom_tyler created CBD-24 forEnable bi-directional stream-spec/*.cbdsst workflow.
Provide a workflow that allows developers to modfiy CBD configurations
via traditional stream s...pec edits, or via changes to the *.cbdsst file.
A change in either updates the other implicitly via trigger actions.
This change will help make CBD automation somewhat more transparent.
Implementation Detail: Presently, the CBD trims version specifiers
from the versioned form of the stream spec (the *.cbdsst file). This
is done for a few reasons:
* Earlier versions of P4D didn't support version specifiers in
stream specs, so they had to be trimmed off or the server would
reject them. P4D 2015.1 supports changelists only as revision
specifiers.
* CBD supports any valid revision specifier, while P4D 2015.1
supports only changelists. P4D 2015.2 may add support for
automatic labels and timestamps. That leaves only static labels
(important) and workspaces (not so import) as revision
specifiers supported by these CBD scripts outside of P4D.
Decision:
* CBD automation will detect P4D version:
o If P4D < 2014.1, strip all revision specifiers going from
*.cbdsst -> live stream spec.
o If P4D < 2014.2, strip all revision specifiers going from
*.cbdsst -> live stream spec except changelist numbers.
o If P4D < 2015.2, strip all revision specifiers going from
*.cbdsst -> live stream spec except changelist numbers and
automatic labels.
Warning: The ability to update stream specs directly had proven a
helpful workaround to earlier CBD issues which have since been
eliminated. This change would eliminate that flexibility, though it
seems it is longer be needed. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-23 forImplement flexbile feature toggles.
Flexible means feature can be enabled globally, per-component,
or in some cases more granular, e.g. per-stream.
...
Example: Make the versioning of the Remapped field
(CBD-17) be configurable as always on, always off,
or always on but limited to the main branch. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-22 forReject submit of invalid *.cbdsst files.
Reject submission of *.cbdsst files that are improperly formatted in some way.
The user should get the subm...ission rejected with a warning (the more specific the better)
describing the error. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-6-3 for | |
|
|
| 8 years agosuper-tom_tyler created CBD-6-2 forSubtask: Define Key Structure for Classic keys.
Streams uses cbd_stream_* and a defined format. Classic
needs a similar defined format for the... REWRITE logic to
work. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-6-1 for | |
|
|
| 8 years agosuper-tom_tyler created CBD-6 for Complete implementation of CBD for Classic. | |
|
|
| 8 years agosuper-tom_tyler created CBD-3 forAdd support for Task streams.
At issue: In SSTemplateUpdate.py, the 'p4 describe' output for task streams
doesn't list depot files (unless run from... the Stream workspace associated
with that task stream). This breaks upon creation of a new Task Stream.
It doesn't prevent creation of the Task Stream, but fails to update CBD
keys for that task stream. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-21 forKeys not cleaned up when import or shore is removed from *.cbdsst.
Removing an import or share in the *.cbdsst leaves the keys hanging around.
"Blob...bing" the CBD keys (See CBD-16) should solve this issue, but I'm
filing this bug to remember to verify that. « | |
|
|
| 8 years agosuper-tom_tyler modified job000248 forKeys not cleaned up when import or shore is removed from *.cbdsst.
Removing an import or share in the *.cbdsst leaves the keys hanging around.
"Blob...bing" the CBD keys (See CBD-16) should solve this issue, but I'm
filing this bug to remember to verify that. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-21 forKeys not cleaned up when import or shore is removed from *.cbdsst.
Removing an import or share in the *.cbdsst leaves the keys hanging around.
"Blob...bing" the CBD keys (See CBD-16) should solve this issue, but I'm
filing this bug to remember to verify that. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-21 forKeys not cleaned up when import or shore is removed from *.cbdsst.
Removing an import or share in the *.cbdsst leaves the keys hanging around.
"Blob...bing" the CBD keys (See CBD-16) should solve this issue, but I'm
filing this bug to remember to verify that. « | |
|
|
| 8 years agosuper-tom_tyler modified job000247 forExtra whitespace in cbdsst isn't handled perfectly.
In the following examples, pretend '=' is a space ' ':
If you have 'share==...'
It will... screw up the stream spec, it needs to be 'share=...'.
With python we could probably make that a bit more 'friendly' and strip extra whitespace problems. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-20 forExtra whitespace in cbdsst isn't handled perfectly.
In the following examples, pretend '=' is a space ' ':
If you have 'share==...'
It will... screw up the stream spec, it needs to be 'share=...'.
With python we could probably make that a bit more 'friendly' and strip extra whitespace problems. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-20 forExtra whitespace in cbdsst isn't handled perfectly.
In the following examples, pretend '=' is a space ' ':
If you have 'share==...'
It will... screw up the stream spec, it needs to be 'share=...'.
With python we could probably make that a bit more 'friendly' and strip extra whitespace problems. « | |
|
|
| 8 years agosuper-tom_tyler created CBD-19 forSupport depots with multiple modules.
This is a code research and testing task. CBD was originally developed
with the assumption of a 1:1 rela...tionship between stream depots and
components.
This task is to ensure that customers using one stream depot to contain
multiple component do not experience any adverse effects with CBD. « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-18 forAdd Windows support for server-side CBD back end.
While the broker & trigger back end implementation works with any and all client platforms, it... only works when the back end is Linux. Add support for Windows only environments (and of course do it in a way that doesn't break existing Unix/Linux platform support). « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-18 forAdd Windows support for server-side CBD back end.
While the broker & trigger back end implementation works with any and all client platforms, it... only works when the back end is Linux. Add support for Windows only environments (and of course do it in a way that doesn't break existing Unix/Linux platform support). « | |
|
|
| 8 years agosuper-tom_tyler modified CBD-17 forVersion 'Remapped:' and 'Ignored:' fields.
The initial CBD implementation updated only the 'Paths:' and
'Description:' fields in the live Stream Spe...c when the versioned
form, the *.cbdsst file, was updated. The 'Remapped:' and
'Ignored:' are ignored.
That functions as initially designed, based on the "miminally
intrusive" philosphy to avoid unnecessary impact of the custom
CBD logic. However, later it was deeemd better for all aspects
of the stream spec to be dictated by the versioned from of the
file if we're going to have one at all, so they can go through
the same merge down/copy up flow as Paths updates. « | |
|
|
| 8 years agosuper-tom_tyler modified job000244 forVersion 'Remapped:' and 'Ignored:' fields.
The initial CBD implementation updated only the 'Paths:' and
'Description:' fields in the live Stream Spe...c when the versioned
form, the *.cbdsst file, was updated. The 'Remapped:' and
'Ignored:' are ignored.
That functions as initially designed, based on the "miminally
intrusive" philosphy to avoid unnecessary impact of the custom
CBD logic. However, later it was deeemd better for all aspects
of the stream spec to be dictated by the versioned from of the
file if we're going to have one at all, so they can go through
the same merge down/copy up flow as Paths updates. « | |
|
|
