After some more testing it does the appear that the assumption that a scheduled group profile DOES NOT display a second icon in the notification area whereas a scheduled non-group profile DOES display a second icon is true.
If you only click on this second icon to display the UI then this instance will exit as normal when scheduled profile finishes, however if you perform some operation such as displaying a log file then it will not exit after the scheduled profile finishes.
When you have one of these orphaned instances left hanging around then one of following appears to happen, when ANY profile subsequently runs. (This can also happen without any orphaned instances, see below.)
1. It runs correctly.
2. If a profile that completes very quickly (such as copying a single small file triggered by ‘On Changes’) then it may run twice, however both times it runs it behaves as expected. But it does appear that both instances ran the profile.
3. A ‘Unknown Error’ status appears in the results column while the profile is running but then when the profile finishes it changes to ‘Success’ and there is nothing in the log file to indicate a problem. The profile can start normally, with ‘Running’ in the status column, with the ‘Unknown Error’ appearing sometimes several minutes after the profile has started. Only one icon is being animated and there is no obvious indication that the other instance of SyncBack is doing anything. If the profile is part of a group profile then the results column for the group will also contain ‘Unknown Error’ while the profile is running and this will remain after the all the profiles in the group have completed, leaving the paradox of the group profile indicating there was a error but the profiles within indicating that everything is OK. It is concerning that if SyncBack encountered some sort of problem to cause it to display the error that it does record this in the log file and completes with a success status.
4. The profile fails, with an ‘Abort’ Success. I have only seen this occur with a group profile. I my case I have a group profile containing two profiles set to run sequentially, both fail with a ‘Abort’ status in the results column. If we call the two profiles A & B, in A’s log file the result is recorded Aborted, further down in the log the reason for the abort is given as ‘<name of B>(<group name>): The profile is already running or is due to run’. The log file for B is slightly different in that it has the message ‘<name of A>(<group name>): The profile is already running or is due to run’ as the result. The point that may be of interest is the log for A refers to B and B refers A. This is the least frequent case and although it has occurred several times it has only occurred once when I was actually looking at the UI, in this case as soon as the group profile started the status of the first profile was ‘Unknown Error’ for about a second then in changed to ‘Abort’, the second profile then followed the same pattern.
All this raises the question does any of this happen without an orphan instance of Syncback, the short answer is YES. In normal operation if there are two (or more) instances of SyncBack running, one running in the background to run periodic, on changes and on insert profiles, and one (or more) running scheduled profiles, then if ANY other profile starts then it may encounter the issues described above. This is less frequent than in the orphan case, but it does happen. I have seen and been able to reproduce the ‘Unknown Error’ case (see below), but I haven’t been able to reproduce the ‘Abort’ case at will (unless I deliberately create an orphan instance).
I hope this all make some form of sense.
Andrew.
If you only click on this second icon to display the UI then this instance will exit as normal when scheduled profile finishes, however if you perform some operation such as displaying a log file then it will not exit after the scheduled profile finishes.
When you have one of these orphaned instances left hanging around then one of following appears to happen, when ANY profile subsequently runs. (This can also happen without any orphaned instances, see below.)
1. It runs correctly.
2. If a profile that completes very quickly (such as copying a single small file triggered by ‘On Changes’) then it may run twice, however both times it runs it behaves as expected. But it does appear that both instances ran the profile.
3. A ‘Unknown Error’ status appears in the results column while the profile is running but then when the profile finishes it changes to ‘Success’ and there is nothing in the log file to indicate a problem. The profile can start normally, with ‘Running’ in the status column, with the ‘Unknown Error’ appearing sometimes several minutes after the profile has started. Only one icon is being animated and there is no obvious indication that the other instance of SyncBack is doing anything. If the profile is part of a group profile then the results column for the group will also contain ‘Unknown Error’ while the profile is running and this will remain after the all the profiles in the group have completed, leaving the paradox of the group profile indicating there was a error but the profiles within indicating that everything is OK. It is concerning that if SyncBack encountered some sort of problem to cause it to display the error that it does record this in the log file and completes with a success status.
4. The profile fails, with an ‘Abort’ Success. I have only seen this occur with a group profile. I my case I have a group profile containing two profiles set to run sequentially, both fail with a ‘Abort’ status in the results column. If we call the two profiles A & B, in A’s log file the result is recorded Aborted, further down in the log the reason for the abort is given as ‘<name of B>(<group name>): The profile is already running or is due to run’. The log file for B is slightly different in that it has the message ‘<name of A>(<group name>): The profile is already running or is due to run’ as the result. The point that may be of interest is the log for A refers to B and B refers A. This is the least frequent case and although it has occurred several times it has only occurred once when I was actually looking at the UI, in this case as soon as the group profile started the status of the first profile was ‘Unknown Error’ for about a second then in changed to ‘Abort’, the second profile then followed the same pattern.
All this raises the question does any of this happen without an orphan instance of Syncback, the short answer is YES. In normal operation if there are two (or more) instances of SyncBack running, one running in the background to run periodic, on changes and on insert profiles, and one (or more) running scheduled profiles, then if ANY other profile starts then it may encounter the issues described above. This is less frequent than in the orphan case, but it does happen. I have seen and been able to reproduce the ‘Unknown Error’ case (see below), but I haven’t been able to reproduce the ‘Abort’ case at will (unless I deliberately create an orphan instance).
I hope this all make some form of sense.
Andrew.