New Members
  • Content Count

  • Joined

  • Last visited

About mlwang

  • Rank

Recent Profile Visitors

370 profile views
  1. Finally! Love the new per-folder issue notification. Thanks!
  2. scegg beat me to it. I've just upgraded to High Sierra (10.13.2), and this is quite annoying.
  3. @Helen, Got home early thanks to the Typhoon. The debug log has been sent to support as you instructed. Turns out there's no predefined hosts for any folder on the device "carlo" (the mac desktop where the screenshot was taken). The "folder_defaults.known_hosts" setting in Power user preferences is empty, too. There are only 4 "Power user preferences" using non-default values: folder_rescan_interval - 18000 send_statistics = false enable_journaling = false peer_expiration_days = 1 In the meantime, a new ghost has appeared, so now the device "carlo" is sync'ed to 3 other "carlos."
  4. Ah, Ok. Then the answer is yes, I'm very sure btsync (2.3.8) is gone. And as I said, I started Resilio Sync (since RC1) with a new identity, and switch all devices sharing the same identity with carlo to the new identity after upgrading to 2.4 RC1. So even if btsync is still running secretly on carlo (or any other device, for that matter), the two should not talk to each other, right? I would be very worried if they do. I'll do as you instruct and collect the logs tomorrow night (Taiwan time, +0800). I'm just home and it's a big day at work tomorrow, so I need the rest of the day to prepare for it. Sorry.
  5. @Helen, sorry, but I don't understand what you mean by confirming btsync is not running on carlo peer. As I said, the screenshot was taken on carlo, so of course btsync was running on carlo. The problem is, there is only one carlo I know of, so where do the other two carlos come from? Carlo should not be listed as a peer of itself, let alone twice. All my devices (same identity or not) use a unique listening port (the one configurable in Advance Preferences, just to be sure we're talking about the same thing). Yes, I do use predefined hosts on some of my devices, but not sure if carlo uses the feature. I'll check and get back to you when I get home (carlo is at home). As to debug logs, there's none since turning off debug logging is the first thing I do when installing btsync. Will they still be useful if I turn debug logging on now? If so, for how long do I need to leave the logging on before collecting the logs?
  6. I'm seeing this as well. I first noticed the phenomenon after upgrading to v.2.3.8. I cleaned out everything and started over (with a new identity) after v.2.4 RC1, and the phantoms disappeared for a while. Now I'm on RC2 and the ghosts have come back. Not sure when they started to appear again (RC1 or RC2) though. Please see the attached screenshot below for an example. The screenshot was taken on my main desktop named carlo. It's the only computer named carlo in my possession. As shown in the screenshot, however, Resilio Sync insists it's synced to two other computers also named carlo. Edit: forgot to mentioned that I'm on macOS 10.11.6, using Sync 2.4 RC2. So are all the other nodes sharing the same identity, as well as another desktop using a different identity (for I don't want it to have access to everything).
  7. @RomanZGot it. Thanks! Will wait for the next version.
  8. @RomanZ Understood. Thanks! It's important to save the RW key, then. Good to know. One more question, if I may: got the following message when trying to set up my first Encrypted folder: "Selected folder is already added to Sync." The message is plain, but the limitation is puzzling: if an folder is important enough for me to save an extra copy on an unsecured host, chances are it's already sync'ed to other secured hosts, isn't it? In other words, sync'ing to encrypted folders is only good for people who have no trusted hosts to sync to. Not that it's useless this way, but much less useful than it could be. Still, thanks for taking the time to explain all this.
  9. Sorry, I don't understand. Let's borrow bruto's scenario above: You mean if A dies, I could decrypt the files on B with the RW key of the share? How do I do that? And if I can do that, why can't I do it on the files in the Archive on B (to restore accidentally removed file) when A is still alive?
  10. I'm glad I found this thread before setting up my first encrypted folder. A question: according to the help page for "encrypted folder", I like that, but how am I going to put the backup to use should a disaster strike--when, e.g., a file is accidentally deleted and the encrypted copy in the Archive folder on the untrusted machine is the only copy left--if I can't recover the file by putting it back? Could someone please give me an example where the encrypted backup can be useful? Thanks!
  11. The discussion about pricing model can't be separated from that on price level. The option to buy a $20000 new car with 4 yearly payments of $5000 is called an installment plan and is usually welcome. The option to buy a $20000 new car with 4 yearly payments of $20000 is called insanity. The same applies to btsync. While some might object a subscription model on principle, most will probably consider it acceptable, I suspect, if the price is set at, e.g., $10 a year. The problem is, $40 is pretty much a price many consider to be about the right price, as a one-off payment, for such software. But btsync is not a car; it's in a much younger and much less well-defined market. Its closest competitor feature wise is an open-source freeware. How do we decide the "right price" for btsync? Behavioral science suggests most people compare it to the market leader in the same or an adjacent market. In this case, it's probably Dropbox. Then people deduct the perceived extra cost associated with Dropbox's extra service (the online storage) to get the answer. That may or may not be fair, and certainly btsync is entitled to test the market with a price it considers to be fair. This will work only for people who consider the current price level is fair (acceptable), and only when "perpetual" means free supply of "all future updates". It would be more attractive if the "future" is better defined, and if people can be assured that the company won't come out with BTSync Advanced (e.g.) and say it's a new product that requires new payment/subscription. Not an easy task I must say, given the company's current image.
  12. While the company should get some credits for not turning a deaf ear, and for staying tolerant when competitor products' names are mentioned here, your answer is incomplete and not exactly convincing. It's incomplete because you didn't address the 10-folder limit flip-flopping issue, but I'll leave it. This is really not logical. If you truely want to "continuously improve the software and offer quicker releases," you should use the regular software licensing model: charging for new purchases and major upgrades (preferably at a discount for the latter). With the subscription model, you get paid the same amount no matter how many improvements you make to the software in a year. The only incentive for "quicker releases" is to attract new subscribers. With the regular software licensing model, "quicker releases" will not only get you new users, but also new revenue from existing users. Which is more powerful should be apparent. Software as subscription, when no substantial resources or other bona fide services are involved, is therefore an admission that one is not really planning major improvements down the road, but still wants to keep getting paid anyway. A company can do that when 1) it has a compelling product that has no peer in the market, or 2) it has a lot of current customers and it's costly for them to switch (in other words, they're trapped). The second arguably applies to Microsoft (for Office 365), but probably not to btsync. So I guess you think you have a compelling, peerless product. Time will tell if you're right, but I guess the company is a little sheepish in taking this position, for otherwise you don't have to impose the 10-folder limit, which is there for no technical justification but to make the Pro version more "compelling". One other thing: the traditional licensing model is better at reaching customers in new economies. US$40 might be nothing to some, but a lot for people in the 3rd world. The traditional licensing model makes it possible for many to buy software they need because they don't have to keep pace with every paid upgrades. I do that with operation systems and many other applications. With the subscription model, I must be sure I can afford it in the long run, and more often than not, I end up not subscribing in the first place. But perhaps you just don't care, and it's certainly your freedom not to. There's very little I can do except to stay on the free version (for that I'm truely thankful) for as long as I can, to stop promoting btsync, and to keep looking for alternatives. ps.: just saw several new posts and some of my points have been made, but since I've composed it, I'm still posting it.
  13. This has to be one of the most puzzling changes made in 1.4 betas. While the old btsync properly identified such files with their full name (including their base name, i.e.), now btsync shows you only the name of the extended attributes, making it impossible to tell, e.g., which file's resourcefork is stuck, without consulting the log. See the screenshot below for an example: