Jump to content


  • Posts

  • Joined

  • Last visited

Profile Information

  • Gender
    Not Telling

FileMaker Experience

  • Skill Level
  • FM Application
    16 Client

Platform Environment

  • OS Platform
  • OS Version
    10.11 el captain

FileMaker Partner

  • Certification
    Not Certified

Recent Profile Visitors

3,976 profile views

sfpx's Achievements

Community Regular

Community Regular (8/14)

  • First Post
  • Collaborator
  • Conversation Starter
  • Week One Done
  • One Month Later

Recent Badges



  1. Looking at the log I can see that the problem occurs when the recovery mode is activated. This keeps happening for many users causing ultra long syncs. This happened this morning again. What is the recovery mode ? What cause its activation ? How can we avoid it ?
  2. I pretty much can assure you those records were not updated on the server prior to the sync. The sync was actually updating both the server and the device records. So after the sync, all his records were updated on the server (with timestamps in the sync period) I will send you the logs
  3. I'm checking a sync of a user and mirrorsync is updating all his records. It's like Mirrorsync completely forgot what was synced or not and is forced to check every record, It's extremely long. Any known cause for this ?
  4. Some more thoughts 1.I tried your privilege set idea in addition of the spoke "willdelete" callback. I made sure to update the records that were going to be deleted (unsuccessfully because of of the privilege set) to force them to be resynced on the server on the next sync. It works. 2.This solution will prevent deletion of new created records between the last backup and the database corruption and that's great but we would still lose the changes made to records between the last backup and the db corruption (there would be modified records on the device that would not be changed on the server). So I guess , to not lose anything I also need to either backup mirrosync each time I backup the db or reset mirrorsync sync data on the server to force a long sync similar to an initial sync with pre populated data. If want to backup mirrorsync data , how do I proceed to do it when fms backup the db ?
  5. Well, if the sync only sends changes from the device to the server then I can't have changes made on webdirect sent back to the device. My conflict resolution is set to "Most recent wins'. But it does not matter. The records are marked as synced and they are no longer on the server after the backup upload so the records will be deleted on the device. ADDED: I don't backup mirrorsync. Should I backup mirrorsync each time I backup the DB and restore both at the same time ? Would that fix everything then ? I like this idea. I will try it.
  6. Can I use the spoke-level "will delete" callback to avoid deletion for certain tables ? Let's say the user creates a record on his device. He syncs at 1PM. The database gets corrupted at 1:20 PM. We have to revert back to 12:45PM backup He syncs again at 3PM. MS will try to delete the new record on the device because it's flagged as synced and is no longer on the server. In the "will delete" spoke callback, we cancel the deletion by omitting the record. Would it work ? Would it be synced back to the server or do we need to modify the record so the timestamp changes ? In the end, I want a bi-directional sync that never deletes anything on the device (at least for certain crucial tables).
  7. We had a database corruption and had to revert to a backup last week. Some users had synced new data after the backup so when they synced after the it was uploaded, their data was deleted on the device. What I wanted to do is to set the conflict resolution to "spoke always wins" so when mirrorsync does not find the record on the server it syncs it again on it. To my surprise it was not an option. Of course I can set the sync to be unidirectional but I would lose some functionality on webdirect. What's the solution ? Mirrorsync config ? Customization script ? Should I backup mirrorsync at the same time as the database backup and restore everything at the same time? What would be the impact of this on the following syncs ?
  8. It's ms 4 I know it's an older version and we plan to upgrade in the upcoming weeks/months but we never had this problem before and out of the blue it has started doing that this week. I'm quite worried about it and I would like to understand what can cause this (not sure I understand the error message) By looking at the logs it happened on about 2-3 tables for 2 users so far. Is this a known bug of ms 4 that has been fixed in a more recent version ? Anyway ms 4 still works well and we never had this error before.
  9. Started to have this error out of the blue this week. This resulted in record deletions on the device (thankfully not on the server). I had to force full deletion of the app + initial sync for 2 users so far. This is quite a huge bug. What can cause this exactly ? We haven't touched anything related to the tables affected.
  10. A little more info about the issue When the pdf file is inserted in the device you can see this info in the container attributes [General] Filename: .pdf Storage Type: Embedded MD5: FAE885A65D413D76CE03826013DCA437 File Size: 24038 Internal Size: 235318 External Size: 0 External Files: 0 [Image] Width: 792 Height: 612 If I take the .fmp12 from the device and load it in filemaker pro, I still can see the same attributes and the container preview. On the server, after a sync, the container [image] attributes are no longer there and the container will not preview the pdf file . I guess it's related to the fact that mirrorsync is using base64 instead of the filemaker insert file function. I would say that it's more a filemaker problem than a mirrorsync one. Now, if someone has an idea how to get back those attributes I would not mind adding code post sync to fix this.
  11. User inserts pdf on the offline device. The container is set to Optimize for Images User can see the first page of the pdf in the container (as an image) After the user synced, the same container field on the server will display the file name instead of the image of the first page. Is there a way to fix this ? P.S. Mirrorsync 4
  12. We are currently using mirrorsync 4 with 50 devices. We wish to update to mirrorsync 6. When we tried the upgrade button on the order page we were told our license expired and can not be upgraded. Does that mean we must pay for another 50 devices ? What's the process of updating from an expired license (with devices) to mirrorsync 6 ?
  13. Hi Wim, We had a bit of discussion about this in the filemaker forums. The database can be opened using the "option" command. Filemaker then "repairs" the db. No recover needed.
  14. I used mirrorsync and fms17 without any database corruption for a full year (and fms16 for a year too) Since I upgraded to fms 18 the database got corrupted twice. Both times there was the sequence of events like this example of yesterday 9:48 PM a user seems to cancel a sync User canceled action (JDBC URL: jdbc:fmp360://mysite/mydb, SSL: false / SQL statement: SELECT _gSync1, lastErrorMessage FROM MirrorSync WHERE -script='MirrorSync (HeliCloud)' AND -script.param=? / SQL params: [doServerAction One minute later the user retried Error 802: Unable to open file - Check to make sure that the database file is hosted on FileMaker Server In the events.log file 2019-10-16 21:48:49.848 -0400 Error 618 mysite Database or temporary file "filewin:/C:/Windows/Temp/FMTEMPFM2932_5.tmp" is damaged and has been closed. (4294967295) 2019-10-16 21:48:49.848 -0400 Error 681 mysite Closing database "mydb" due to damaged temporary file... Any idea of a way to fix this ?
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.