Skip to main content
A
Anonymous
July 7, 2009
Question

Synchronization issues with RoboHelp 8 and RoboSource?

  • July 7, 2009
  • 1 reply
  • 1681 views

We recently converted from RoboHelp 7 to RoboHelp 8 and have noticed a rise in synchronization issues, particularly issue involving the root.fpj, since we converted. In some cases, authors open their projects to find "X"s through the topics they checked in last night. In another case, an author opened her project to only find 10 topics remaining - we had to revert to a previous fpj because the one in her project was 3/4 gone!

Has anyone experienced this? It seems to be happening much more now that we are on 8. It seems to happen most frequently when authors check in 1 file at a time, instead of using the check in all button. Why is this happening, and what can I tell my team to do to prevent it?

This topic has been closed for replies.

1 reply

Gravenstein
Gravenstein
Inspiring
July 9, 2009

Yes, we run into this, too, although we are still using RH6. I don't know what causes it; we just fix the problems as they come up. The problem is worst with changes at the root level.

G

A
Anonymous
July 15, 2009

Thank you for your reply. When you say the problem is worst at the root level, does this mean that you store topics within folders in your project? Where do the bulk of your HTM files reside? At the root, or in another topic folder? And if they are in another topic folder, does that help with this problem?

We had it happen again yesterday - same issue - the root.fpj deleted all but about 10 entries. Then RoboHelp decided to jump on board and hide the remaining topics in RoboSource, so they appeared missing from the repository and peoples' projects until I made them visible.

I am trying to trace the common behavior that triggers this, because even though it is fixible, it consumes time to troubleshoot and resolve. The only common thing our two project teams did that I could identify was rename topics - something we know RoboSource doesn't enjoy, but should support.

Have you found any behavior or actions that seem to trigger this behavior?

A
Anonymous
July 16, 2009

Hi. I work with Gravenstein, and she's out of town this week, so I thought I'd throw out a couple of notes.

Yes, nearly all of our topic files are in folders, and it does seem to help. We have one folder per project to hold Captivate demos and the topics that display them, which Rick Stone was talking about just yesterday on the other forum.

Renaming files and moving files between folders create problems because RoboSourceControl has to create the new file, copy in the content, save the event in the file's history, and delete the old file. If the network or server hiccups during this lengthy process, it is likely to crash RH and leave weird unsynchronized pieces in RSC, as you've seen. This is most likely to happen in the afternoon, when server traffic is highest.

For prevention, limiting the practice of renaming is a good start. Allow as few topic renames as possible, and restrict them to low-traffic times on the network or server. Create topic naming conventions and enforce them brutally.

Down the road for prevention, I think it would be worth your while to set up a folder structure. Name your folders something innocuous, and never rename folders while the project is under source control. If you absolutely must, create new folders and move topics into them. If at all possible, take the project out of RoboSourceControl when you have time to do the heavy roadwork. Big changes are best made out of source control - that includes wholesale name changes and structural changes like moving files into folders.

For mitigation (not prevention), have developers make .zip backups of their project source files when they close the project for the day. These backups are handy when the .cpd (the RoboHelp project "definition" on each dev's PC) and RSC (the saved project files that don't include the .cpd) disagree. I'm thinking of those .fpj and .hhk files, mostly.

I'm afraid that's a lot of words and not much help here, but it might lessen the crash frequency enough to get a handle on any unique factors in your environment that might be causing trouble.

Oh, yeah - and all the traditional crash-prevention strategies: making sure you don't have inappropriate files in RSC (.cpd, .hhp, .pss - RH8 may have more types that don't go into source control), using Check In to save the topic you're working on before you move to the next topic (Save All is a good frequent flyer, because it updates the .cpd, but it doesn't always play well with RSC)...

HTH,

Elisa

Terms & ConditionsAccessibility statement
Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices