Jump to content


  • Content count

  • Joined

  • Last visited

  • Days Won


DavidHertzberg last won the day on April 12

DavidHertzberg had the most liked content!

Community Reputation

37 Excellent

About DavidHertzberg

  • Rank
    Occasional Forum Poster

Profile Information

  • Gender
  • Location
    New York, NY
  • Interests
    Retired applications programmer, with a few Macs at home.

Recent Profile Visitors

1,046 profile views
  1. I have for several years been doing what it says in this old post, but now using the Desktop Edition of Retrospect Mac 14.6.0 on the Mac Pro "backup server"—which is running under macOS 10.12.6. In fact since yesterday I have been doing the same thing using a test-build of Retrospect Mac 15.1.1, supplied to me by Tech Support so that I can help pinpoint the -530 Bugs. One "client" is a MacBook Pro running the latest production Client (14.6.0 until yesterday, now 15.1.1). The other "client" is still a Digital Audio G4 running the 6.3.029 Client under OS X 10.3.9; as stated in the linked-to post, I have to change that Client's switch from Off to On. Once a week I also back up a an additional local drive on the Mac Pro that belonged to my late friend; I've left it as is with OSX 10.6—although I rarely boot from it, because my friend used a lot of PowerPC applications. As the head of Retrospect Tech Support states here, you shouldn't be using Client 6.3.029 to backup "clients" booting under anything past OS X 10.4. However a year ago I did some older-version-of-application selective Restores on my Digital Audio G4, both to its OS X 10.3.9 boot drive and to one of the OS 9 drives also installed on that "client".
  2. DavidHertzberg

    Shut Down Computers Automatically

    Not any more there isn't. There was an option to shut down a client after completing a backup; that option is described on page 228 of the Retrospect Mac 14 User's Guide, in the "Legacy Client" appendix. However it is not shown in the screenshot on that page, and the description refers to the Special Menu, so I think it is only available if the Client is running on an OS 9 machine.
  3. DavidHertzberg

    No more instant scan on MacOS?

    suyashdruva says in the last 21 May 2018 post in that Apple Developer Forums thread "I observe case and normalization are preserved for operation other than delete. So why behavior is different for file/folder delete event? It seems while accumulating events, paths are getting normalized except this case." CrashPlan had (and evidently still has) a File System Scan that was scheduled to run by default at 3 a.m., because it "Requires more resources". CrashPlan used to use Spotlight instead of FSEvents—which Retrospect uses, so that section of the page used to say "Spotlight does not report deleted files in real-time, so CrashPlan only detects deleted files during the scheduled scan". Maybe Retrospect will have to adopt the same approach. P.S.: insont got in his immediately-preceding post while I was researching the second paragraph for this one. I think what I've just said still is applicable. P.P.S.: For the masochists and/or frustrated systems programmers among you, here's an applicable blog post—with comments—on APFS filename normalization.
  4. This over here is the new thread. Since it contains much more current information, let's use it.
  5. DavidHertzberg

    No more instant scan on MacOS?

    The OP in the 21 May Apple Developer Forums thread I linked to in this previous post used the "handle" "suyashdruva", so I just did a little Googling for "suyash druva". It appears that his name is Suyash Singh, and he works for a company named Druva. And a look at an overview video on Druva.com confirms they have product(s) that do AWS cloud backups including for "endpoints". Reading some of the references in this section of the Wikipedia "Backup" article has made me aware that "endpoint" is industry slang for end user devices such as PCs, laptops and mobile phones. And that would include Macs, so the APFS problem insont/martin reports sounds as if it's Mac-backup-industry-wide and not just limited to Retrospect. So "Walnut Creek, we have a problem"; sorry for my initial skepticism, Martin. BTW, even the newest of my 3 Macs is on macOS 10.12.6, and they use good old HFS+. So I'll let Instant Scan continue to shorten my daily incremental backups of one Mac by 8-10 minutes.
  6. DavidHertzberg

    No more instant scan on MacOS?

    insont/martin, I doubt your veracity because: [1] No other threads, other than the ones you have posted in, have reported a problem with APFS and Instant Scan for Retrospect Mac except for this one back in October 2017. [2] Retrospect Engineering has lately given fair warning about planned feature changes that would affect some administrators adversely; they put yellow-flagged Notes about two such changes (once of which has already been expanded into a Knowledge Base article) in the cumulative Release Notes for Retrospect Mac 15.1 (and one of those was already in there for Retrospect Mac 15.0)—so I find it a bit difficult to believe that Retrospect Engineering (which IME communicates with customers only through a Support Engineer) would announce such an adverse change directly to you. [2] is not impossible to believe; the next-to-last substantive paragraph in this post suggests that IMHO the "real" engineers are being given unprecedented freedom in documentation. Also, the Apple Developer Forums thread linked to in my preceding post indicates there is a problem with the APFS change in "normalization" of file names. Finally, this would be such a massively adverse change that Retrospect Product Management may have "chickened out" on announcing it, which would explain why you "had a week-long exchange with Retrospect support" and why this Knowledge Base article has not been updated. As a mere customer, I am not allowed to view Support Cases other than my own. However I will file a Support Case, referencing this thread and Support Case 61949—which Retrospect Support has the capability of viewing.
  7. DavidHertzberg

    No more instant scan on MacOS?

    insont, The only 2018 information I could find about this is this thread on the Apple Developer Forums. Forgive me for asking this, but I notice your profile contains no information about you, that you joined the Forums on 30 June, and that your profile says "The recent visitors block is disabled and is not being shown to other users." Is what you are saying you heard from Retrospect Engineering true, or is it "fake news".
  8. If you're on Retrospect 15.1, you may still be entitled to free personalized Retrospect Technical Support. Here is why and how to file a Support Case. If you are in the U.S. or Canada, I would also suggest making a phone call to R.T.S. at (888) 376-1078 or (925) 476-1030 between 8 a.m. and 4 p.m. Pacific Time. If you are in Europe (I can't tell; please update your Profile with at least your country of location), from what I've seen on these Forums it sounds as if the phone support there is not very knowledgeable.
  9. What version of Retrospect Mac are you using? If it's later than Retrospect Mac 12, Fast Catalog Rebuild has been removed—see the first paragraph of this post (and mentally substitute Retrospect Mac 13 whenever you read Retrospect Windows 11 and Retrospect Mac 14 whenever you read Retrospect Windows 12). If you're still using Retrospect Mac 12 or earlier, you might be able to get out of the problem by upgrading—although you can't run recent versions of Retrospect Mac on a PowerPC Mac (which you may be using if you're talking about tapes). If you can upgrade and still mount a compatible tape drive, I would suggest deleting the Media Set and its catalog, creating a new one with the same name, and adding the tapes as members. Actually even if you can't upgrade, I would suggest doing the same thing.
  10. The reason Pete asked bobagentur those questions seems to have had something to do with this post, although it came earlier than speleo14's thread.
  11. DavidHertzberg

    More granular control over email notifications

    LRSFC_DanJ, Glad to see a version of the script hook script you need is already written, so you don't need to go to the lengths I suggested here. Interesting to see that, according to the copyright date, it was written in 2012. This confirms what I had previously learned; script hook scripts, which were announced as a new feature of Retrospect Windows 12, already existed. Maybe they were just new for Retrospect Mac 14, although they didn't work there for Windows Clients until saskia got that fixed over six months later. However, in the "Example scripts" section of this Knowledge Base article, there is a script listed as email notifier Windows specific batch file to illustrate emailing each event. (Old) I just downloaded the .ZIP file linked to at the top of the "Example scripts" section of that article, and the applicable script in it appears to be identical—except for the line-starting plus signs—to what you linked to. That would imply that the script is a Windows batch file, written in a language even more loathsome (looks longingly at old C++ books on shelf)—but more widely known by IT people—than VBA. Could you come up with a combination of checkboxes that would satisfy all administrators, or would individual adaptions of a script be easier?
  12. DavidHertzberg

    More granular control over email notifications

    LRSFC_DanJ, I just remembered that, back in September 2016, I demonstrated that Wake-On-LAN didn't work for scheduled scripts doing Backup of my MacBook Pro—and this was confirmed as a bug by Retrospect Tech Support. I don't use Proactive scripts, so I asked in the immediately-preceding post whether the bug also existed for Proactive scripts. I never got an answer in that thread or the thread that post linked to. The cumulative Retrospect Windows Release Notes have the following entries for 15.1: "Windows Client: Fixed Wake-On-Lan (WOL) for upgraded Windows client (#7358)". "Mac Client: Fixed issue where client did not prevent macOS from going to sleep during backup (#7273)". They also have the following entry for 15.0: "Changed defaults for "Wake-on-LAN" option: disabled for proactive backup and enabled for other scripts (#7237)". Since you're running Retrospect Windows 15.1, you might want to make sure you've updated your Client software on all your Mac and Windows "client" machines to make sure your Mac "clients" are not inadvertently going to sleep during Proactive backups. If your users are intentionally putting Mac "clients" to sleep during Proactive backups, I suggest the Post-it note approach quoted at the beginning of this post. I also suggest that you talk to Tech Support, to find out exactly what the Wake-On-LAN situation is for Clients. BTW I don't see any way any application's "backup server" could wake up a "client" that has been actually powered off, but then I only had high-school physics. Maybe Chinese Intelligence can do it for ZTE cell phones; I bought one about a month ago, so should I be worried?
  13. DavidHertzberg

    PSA: Wikipedia article on Retrospect going away in current form

    The 90-day average of views (article->View History->Page view statistics->Options: Latest 90) of the Wikipedia "Retrospect (software)" article is 25. That's around 20 people viewing it on ordinary days, supplemented by another 15-25 people viewing it on days when there has just been an update. I suspect that most of the extra 15-25 people work in Walnut Creek CA, but the 20 "others" represent potential buyers of the software. Those "others" are the people for whom DovidBenAvraham expanded the article in October 2016. The Wikipedia community is trying to improve the trustworthiness of information in articles. Let's look at the "Backup" article as an example. Having appended a new 2-screen-page section with about 50 discrete references (some used several times), DBA turned his attention on 10 April 2018 to the preceding 7 screen pages of the article—which at that time had 18 references. The preceding pages had not received substantial updates since 2011, and basically represented "community folklore" as of about 2007. The last sentence of the "Hard disk" paragraph in the "Storage media" sub-section said "The main disadvantages of hard disk backups are that they are easily damaged, especially while being transported (e.g., for off-site backups), and that their stability over periods of years is a relative unknown." DBA and I felt that the first part of the sentence was out-of-date for modern portable HDDs, which now have ramp loading and built-in accelerometers. DBA found a manufacturer statement which supported our view, but could not find it again. Meanwhile WP editor JohnInDC found a The Wirecutter 2017 review that covered "rugged" portable HDDs; it seemed to support the old view, but DBA pointed out on the WP article's Talk page that—when read carefully—the review said that the tested drives were fairly shock-resistant although not as shock-resistant as the manufacturers claimed. On 9 June DBA inserted a reference to a YouTube video of an experiment by the German "Timo", which seemed to be a fairly careful limited test of shaking and banging a not-in-a-computer internal Seagate HDD USB-connected to a computer measuring InputOutputs/sec.. JohnInDC rejected the reference, commenting "YouTube is not a Reliable Source". DBA replied on the WP article's Talk page that the warnings about YouTube in the Wikipedia rules referred to copyright violations and one-sided narrator statements, and were not applicable to videos of carefully-conducted experiments—which in his opinion do not include a 2009 mail.ru video of a black-clad demonstrator tossing an un-cabled Samsung portable HDD high over his shoulder and off a wall accompanied by a penny-whistle rendition of the "Imperial March" from "The Empire Strikes Back". There being no response from JohnInDC over the next week, DBA put back in the reference to the "Timo" video. At the same time he deleted the reference to the Wirecutter review, because its 2018 revision has deleted the supporting sentences DBA had quoted. DBA also added a reference to a Toshiba Singapore advertisement he had at last found, which mentioned the ramp loading and accelerometer features of a particular 2018 portable drive model. The preceding two paragraphs describe the current vetting of a technical Wikipedia article. In it DBA represents the ambitious contributor, and JohnInDC represents another editor determined to ensure the accuracy of the contribution. It is worth noting that the updated first 7 screen-pages of the current "Backup" article now have 48 references; DBA's efforts on those pages seem to have encouraged the WP editor Lostraven to add references, many of which are to printed books and journals that Lostraven found on books.google.com. P.S.: JohnInDC deleted the reference to the "Timo" experiment again on the grounds of experimenter "non-reliability". DBA investigated a further "Timo" experiment that destroyed a not-in-a-computer internal hard drive with a hammer, and decided that the first "Timo" experiment was not powerful enough to be worth fighting for. However DBA found a 2007 HGST whitepaper that really justified the improvement in portable drive reliability, and pruned quotes in a reference to it of "buying advice" that JohnInDC objected to. All part of the modern Wikipedia article vetting process, folks.
  14. DavidHertzberg

    PSA: Wikipedia article on Retrospect going away in current form

    If you read my complete post, you'll see my explanation. Good morning to you, but I'm going to bed.
  15. DavidHertzberg

    PSA: Wikipedia article on Retrospect going away in current form

    Sorry, Hofstede, but I've just posted a complete explanation of what has turned out to be the overall topic of the thread. I've been writing the explanation for several hours, but posted the quote from you and the first sentence of a first short paragraph from me as soon as I started writing—just to see how the quoting would work. You evidently replied to the first sentence before I finished my post. I'm in the U.S., not in Germany. Therefore I'm going to catch an abbreviated night's sleep, while you read my complete post. Guten Tag.