PRIORITIZATION AND ACTIONS FROM VERSION 0.3 REPORTS Richard Brooksby, 2000-11-21 1. INTRODUCTION The goal of this document is to "discharge" all the items in the test reports for version 0.3 of the P4DTI, so that those reports cease to be lists of things to do and become reference material. This document will generate the new, trimmed and prioritized list of things to do, which we may enter into a defect tracker. The readership of this document is project staff. This document is not confidential. 2. DECISIONS AND ACTIONS ON ISSUES 2.1. [GDR 2000-11-03] 1. Nice -- minor optimization. 2. Optional -- move delays to start up with a message. 3. Done -- Reported to Perforce by e-mail to support. 4. Done -- Logged for reporting to TeamShare. 5. Essential -- we should at least work around this problem. Underlying cause has already been reported to Perforce. Created job000042. 6. Done -- Logged for reporting to TeamShare. 7. Nice -- could add to P4Win "Tools" menu if we knew the right URL. 8. Essential -- We should provide triggers and enforcement in TeamTrack. Created job000043 and job000044. 9. Ignored. 10. Optional -- submission of new jobs in Perforce. 2.2. [GDR 2000-11-01] 1. Optional -- provide TeamTrack report and Bugzilla query to display outstanding conflicts 2. Ignored -- there's no feasible control over checkout in Perforce, only over submit, which we plan to do something about (see above). 3. Ignored -- reported to Perforce, change unlikely. 4. Nice -- need to add this to the AG. 5. Ignored -- we will do this anyway in our documents. 6. Essential -- must make the replicator a self-starting daemon or service so that it can be managed, and it must re-start sensibly. Created job000045 and job000046. 7. Optional -- document backing out a change as a use case. Suggest linking both the original change and the back-out to the issue. 8. Optional -- tidy up naming conventions 9. Essential -- there should be a progress report, but we should also only replicate issues which are created or modified after a configured date, which is the time of first installation by default, and we should document how to set this time earlier. Most people don't need all their historical issues replicated. Created job000047. 10. Optional -- fix this bug to improve robustness 11. Ignored -- reported to Perforce. 12. Optional -- allow this field to have the same name as in the DT. 13. Essential -- add consistency checking of Perforce and DT configuration, including that the field types correspond. Created job000048. 14. Essential -- same as 13, but check that field names match properly, and normalize case, and document that the integration is case insensitive. Added to job000048 15. Critical -- improve installation and configuration by automatically deducing a lot of configuration information from the DT (already underway). Created job000041. 16. Nice -- purchase a rubber fish and take it to beta test sites as an option 2.3. [GDR 2000-10-31] 1. Essential -- improve quoting of command lines. Created job000049. 2. Optional -- create P4DTI user in Perforce with nice fields 3. Essential -- we definitely need a way to reset as much as possible and re-start the system from scratch, especially if we're generating the jobspec. Of course we can't delete the fixes data, so we can't delete all the jobs completely. Created job000050. 4. Essential -- we must re-read the changes table to make sure that implicit changes like this get replicated back. Created job000053. 5. Optional -- test using plug-ins. 6. Optional -- need to develop a way to tell whether the servers are down and whether the replicator is running, i.e. how up-to-date is the information the user is looking at 7. Essential -- develop reports for QA based on the information gathered by the integration. Created job000054. 8. Ignored -- Duplicate. Added to evidence for job000050. 9. Ignored -- reported to TeamShare. 10. Optional -- document more example configurations. 11. Ignored -- we already recommend job filters in the UG. 12. Ignored -- we can't do much about that. 14. Optional -- we should provide a date field in the e-mail we compose with the local date and time of the machine on which the replicator is running. It can't do any harm, and it might be useful information. 16. Essential -- this situation needs analysis, especially since we currently tell people that they can retrospectively add fixes in the UG. Created job000055. 18. Optional -- earlier enforcement of process would be a good thing, for example, a trigger to work out whether the consequences of a submit will be legal, taking fixes and job status changes into account. 20. Ignored -- already told TeamShare. 22. Ignored -- duplicate. Added to evidence to job000043. 2.4. [GDR 2000-10-30] 1. Ignored -- GDR reported this bug to TeamShare. 2. Optional -- added to job000041. 3. Ignored -- duplicate. Added to job000041. 4. Essential -- we must update the documentation to mention that the P4DTI user needs to be a Perforce super user. Created job job000056. 5. Ignored -- duplicate. Added to job job000041. 6. Essential -- we must escape or translate magic characters in single-select field values. Created job job000057. 7. Optional -- improve the documentation to explain this subtle point. 8. Essential -- we must add this to the AG. Created job000058. 9. Fixed in changelist 4794. 10. Ignored -- already fixed by GDR. 11. Ignored -- it later appeared to work and this report is too vague to do anything useful with. 12. Ignored -- duplicate. Added to evidence for job000041. 13. Ignored -- reported to Perforce. 14. Ignored -- GDR has already suggested improvements to this table to TeamShare. 15. Optional -- this is hard to do and in any case was a workaround for a temporary situation at Mahi. We might increase the priority if someone else asks for it. 16. Essential -- we should tell people about TeamTrack's poor error handling and suggest that they enlarge their event log (or set it to "overwrite as needed" perhaps) and look at it when the replicator reports certain errors. We should update the code to tell them to look at the event log as well. Created job job000059. We've also already nagged TeamShare about their poor error reporting. 17. Optional -- would be less confusing if this were fixed. 2.5. [GDR 2000-10-27] 1. Ignored -- already fixed, says GDR. 2. Ignored -- duplicate. Added to job000060. 3. Nice -- could ping the SMTP server and other servers too, I suppose. But they might not be pingable. 4. Ignored -- TeamTrack GUI aesthetics. 5. Essential -- we must tell users to set this preference in the UG. Created job000061. 6. Essential -- we need to investigate this. Created job000062. 7. Ignored -- already fixed, says GDR. 8. Ignored -- duplicate of problem already fixed, says GDR. 9. Ignored -- duplicate. Added to job000054. 10. Ignored -- TeamTrack flakiness. 11. Optional -- need to at least advise people about this. Replicator goes into conflict at the moment, says GDR. 12. Essential -- the dates must be correct. Created job000063. 13. Duplicate -- added to job000060. 14. Optional -- write data that would otherwise be lost to the log as well as to e-mail. 15. Essential -- we have no documentation of product features in general, and could do with a "product information" sheet, or something like that. Created job000064. 16. Ignored -- TeamTrack problem. 17. Critical -- management problem for RB. 18. Ignored -- it won't be ready. 19. Ignored -- P4DTI-filespecs requirements downgraded and the feature will be removed. 2.6. [GDR 2000-10-26] 1. Ignored -- fixed in latest P4Win. 2. Ignored -- my mistake, not a software problem. 3. Optional -- add how to find ODBC drivers etc. to AG. 4. Essential -- added to job000049. 5. Nice -- improve robustness by checking. 6. Essential -- document jobspec tricks that won't work. Created job000066. 7. Essential -- added to job000056. 8. Optional -- the replicator no longer replicates all changelists. This issue needs to be verified. 9. Optional -- assert more about the output from Perforce. 10. Ignored -- replicator no longer replicates all changelists, just those that it needs to. 11. Optional -- investigate this and try to find out what happened. 12. Ignored -- cosmetic P4Win problem. 13. Optional -- wrap outgoing e-mail messages. 14. Ignored -- duplicate. The replicator stops when it encounters errors it wasn't expecting. This is a problem of error recovery. Added to job000060. 15. Optional -- At the moment the replicator puts things into conflict, which will probably confuse the resolver. But this won't happen with the automatic configuration, only with people who are hacking around. 16. Ignore -- Almost certainly just forgot to enable version control in the TeamTrack user profile. 17. Ignored -- already fixed, says GDR. 18. Optional -- User impact: No way to do resolution from TeamTrack. Could try it. 19. Optional -- improve checking of the configuration. 20. Essential -- log levels will be important for beta testing with remote sites. Created job000065. 2.7. [RB 2000-10-24] 1. Ignored -- already fixed. 2. Ignored -- added to job000041. 3. Essential -- we must document the workflows that won't work well, and this is one of them. Created job000067. 4. Ignored -- already fixed. 5. Ignored -- added to job000060. 6. Essential -- must add procedure. Created job000068. 7. Ignored -- already fixed. 8. Ignored -- already fixed. 9. Essential -- need to write documentation for the resolver in the AG. 10. Ignored -- no, there's no way to do this in the Perforce interface. 11. Ignored -- already fixed. 12. Ignored -- not actually true. We discovered it. 13. Ignored -- just an observation. 14. Ignored -- already told TeamShare. 15. Ignored -- we don't intend to edit fixes from TeamTrack anymore. 16. Ignored -- we don't expect to edit fixes from TeamTrack anymore. 17. Ignored -- we don't expect to edit fixes from TeamTrack anymore. 18. Optional -- GDR made one up that is OK for now. 19. Essential -- created job000070. 20. Ignored -- optimization observation. 21. Ignored -- reported to TeamShare. 22. Optional -- We don't intend to let people edit fixes from TeamTrack anymore, so this can't happen. 23. Ignored -- duplicate. Added to job000063. 24. Ignored -- already fixed and reported in the log. 25. Ignored -- duplicate. Added to job000067. 26. Nice -- it just stays replicated, harmlessly. 27. Ignored -- added note to job000041. 28. Ignored -- cosmetic P4win problem. 29. Ignored -- Perforce are working on it already. 30. Ignored -- not even a good idea. 31. Optional -- duplicate of [GDR 2000-10-26, 18] 2.8. [GDR 2000-10-23] 1. Essential -- we must tell people to uninstall SourceBridge. Created job000071. 2. Essential -- created job000072 3. Essential -- added to job000072. 4. Fixed in changelist 4799. 5. Essential -- created job000073. 6. Essential -- created job000074. 7. Ignored -- already reported to TeamShare. 8. Ignored -- duplicate. Added to job000041. 9-14. Optional -- automatic configuration takes care of most of this, but we'll need to document it for advanced administrators. 15. Ignored -- already fixed, says GDR. 16. Optional -- need to investigate whether this is a real problem for users. 17. Essential -- configuration error could cause serious problem that's very hard for users to fix. 18. Ignored -- unconfirmed, probably not important. 19. Essential -- created job000075 and added to evidence for job000041. 20. Optional -- improve newline handling on blank text fields. 21. Ignored -- fixed, says GDR.. 2.9. [RB 2000-10-18b] 1. Ignored -- there's a readme now. 2. Ignored -- this is just wrong. 3. Ignored -- it's just a note. 4. Ignored -- it's just a note. 5. Ignored -- already fixed. 6. Ignored -- already fixed. 7. Optional -- we should explain what to do if the database looks blank -- that this is probably a licensing problem with TeamTrack. 8. Ignored -- already fixed. 9. Ignored -- already fixed. 10. Ignored -- already fixed. 11. Ignored -- already fixed. 12. Ignored -- already fixed. 13. Ignored -- it's just a note. 2.10. [RB 2000-10-18a] Ignored -- mostly fixed in later build procedure, or insignificant. 2.11. [GDR 2000-10-17b] 1. Ignored -- not very important 2. Ignored -- screenshots were added by GDR. 3. Ignored -- not very important 4. Ignored -- these steps were added to the build procedure. 2.12. [GDR 2000-10-17a] 3-23, 26-29. Ignored -- already discharged by GDR on 2000-10-18. 1. Ignored -- already discharged by RB on 2000-11-20. 2. Optional -- change name of SAG to AG after Bugzilla integration merge. 24. Nice -- improve the message as suggested. 25. Optional -- consider merging "keep" and "replicate" 26. Ignored -- we already changed things to use the visible case ID in TT. [When exactly?] 30. Nice -- improve output format to remove confusion 31. Essential -- clean up and carefully divide assertions from consistency checks and regularize the error recovery code. Created job000060. 32. Essential -- an illegal transition should be backed out completely, including any fixes. ???? Is it already. Ask GDR. 33. Ignored -- we've already documented how to stop the replicator. 34. Nice -- make check script reassuring. 35. Ignored -- not a defect. 2.13. [GDR 2000-10-16] 1-7. Ignored -- already discharged by GDR on 2000-10-17. 8. Optional -- add these problems to the SAG. A. REFERENCES [GDR 2000-10-16] "Integration test report [for release 0.3.0]" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-16. [GDR 2000-10-17a] "Test report for release 0.3.1" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-17. [GDR 2000-10-17b] "Build report for release 0.3.1" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-17. [RB 2000-10-18a] "Release 0.3.2 build report" (e-mail message); Richard Brooksby; Ravenbrook Limited; 2000-10-18. [RB 2000-10-18b] "Test report for release 0.3.2" (e-mail message); Richard Brooksby; Ravenbrook Limited; 2000-10-18. [GDR 2000-10-23] "TeamShare PSG alpha test report, 2000-10-23" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-23. [RB 2000-10-24] "TeamShare PSG Alpha Test Report, 2000-10-24" (e-mail message); Richard Brooksby; Ravenbrook Limited; 2000-10-24. [GDR 2000-10-26] "Alpha test report: Perforce, 2000-10-26" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-26. [GDR 2000-10-27] "Alpha test report: Perforce, 2000-10-27" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-27. [GDR 2000-10-30] "Mahi Networks alpha test notes, 2000-10-30" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-30. [GDR 2000-10-31] "Alpha test report for Mahi Networks, 2000-10-31" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-10-31. [GDR 2000-11-01] "Alpha test report for Quokka Sports, 2000-11-01" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-11-01. [GDR 2000-11-03] "Alpha testing at Quokka, 2000-11-02" (e-mail message); Gareth Rees; Ravenbrook Limited; 2000-11-03. Copyright (C) 2000 Ravenbrook Limited. This document is provided "as is", without any express or implied warranty. In no event will the authors be held liable for any damages arising from the use of this document. You may make and distribute verbatim copies of this document provided that you do not charge a fee for this document or for its distribution. $Id: //info.ravenbrook.com/project/p4dti/doc/2000-11-21/report-actions/index.txt#1 $