saskia

I find that Retrospect sometimes gives a -530 error after having been successful for months and that I can get going again by forgetting and then adding the client again at the server end. A nuisance. There is no change of IP address involved since all of the addresses are statically assigned. I also have clients where the boot partition shows up but not the other partition until I re-install

I use Chrome Remote Desktop between Macs from home to work

Oops. not fixed in release version, but I tested a non-release version a couple of weeks ago that does fix it, so hopefully it will be available publicly soon.

FYI - this bug is fixed in 14.6

Windows client backup events are only recorded when only part of a volume is tagged for backup. Client version appears irrelevant, same behaviour with the 8.2 and 12.5 clients.

It appears the problem is not fixed in the 14.5 release, bug re-opened.

Message from Support: We have reproduced the issue with the external scripting and we have written a bug for the problem. We hope to fix this in our next update.

Retrospect up to date 14.1.0 (138). Handler is barebones simple: >>>>>>>> /Library/Application Support/Retrospect/retroEventHandler.sh: #!/bin/sh tmpfiles="/datalog/tmp/retrochk" num="$#" cmd="$1" all="$@" date "+%D-%T ($num) $all " >> $tmpfiles.log exit >>>>>>> Excerpt from retrochk.log saved by above script: 07/21/17-22:48:19 (4) StartScript Daily 7/21/2017 22:48 /Library/Application Support/Retrospect/intv0013.reh (successful incremental backup of a windows server by this script missing here from log) 07/21/17-22:54:01 (6) StartSource Daily My Computer/OS/datalog /Volumes/OS/datalog/ My Computer /Library/Application Support/Retrospect/intv0016.reh 07/21/17-22:54:29 (18) EndSource Daily My Computer/OS/datalog /Volumes/OS/datalog/ My Computer 5754 5 7 7/21/2017 22:54 7/21/2017 22:54 7/21/2017 22:48 LT-C Normal OS 0 0 successful false

Definitely some backups are not reported.

At this point I am monitoring data passed to retroEventHandler.sh by Retrospect to understand the syntax. BUT unless I am sorely mistaken, some of the successful backups don't seem to be flagged and some of the messages seem to come at the wrong time (StartSource at the end of the backup for example, which is a bit late if your want to make a database quiet, not that I need to). I will grab more info as my usual backups come and go and then update this message. BTW, I have an existing web interface for our computer stats and I will not be using any 3rd party tools to complicate my setup.

So it looks like there is no proper documentation on the arguments passed. So I will have to wing it. Support gave me this link: http://www.retrospect.com/en/support/kb#script-hooks and http://www.retrospect.com/en/support/kb/script_hooks gives a bit of information, but in logging with my own minimal /Library/Application Support/Retrospect/retroEventHandler it is passed a lot of arguments I shouldn't have to guess about... -StartScript Copy Assistant - 2017-07-20, 3:31 PM 7/20/2017 16:36 /Library/Application Support/Retrospect/intv2775.reh -StartSource Copy Assistant - 2017-07-20, 3:31 PM My Computer/Home /Volumes/Home/ My Computer /Library/Application Support/Retrospect/intv2776.reh -EndSource Copy Assistant - 2017-07-20, 3:31 PM My Computer/Home /Volumes/Home/ My Computer 0 0 0 7/20/2017 16:36 7/20/2017 16:36 7/20/2017 16:36 Backup Clients/alpha-dhcp/OS/ada Normal My Computer 0 0 successful true -EndScript Copy Assistant - 2017-07-20, 3:31 PM 0 0 successful for example, what is /Library/Application Support/Retrospect/intv2775.reh it is not a file and for example, what is 0 0 successful true trying to tell me

This is too much email. Am looking into script hooks so I can automate processing with my own bash scripts (running on Mac OSX). I can't find a definitive description of the arguments passed although I am now logging them as a preliminary.

My two schemes are: 1) Proactive backup to tape of all non-cache files less than 100 megabytes in size. This gets all of the important research files which are text anyway. 2) Groomed backup to disk of all files over 100 megabytes. This avoid filling up tapes so fast with big files that tend to be junk but also includes bloated Microsoft Outlook .pst files that might have something useful buried in them. Retrospect will report that a client is backed up if either scheme happens, but I want to be sure that both have been successful recently. What is my least effort way of regularly checking this?

See attached picture. Interface claims that tape set has no members but contains lots of files. There are actually 7 tapes in the set, but it won't show me the info I need. Retrospect 9.0.1 on Mac OS X 10.6.8 Server