Skip navigation
This discussion is archived

Server Connections Interrupted

46277 Views 252 Replies Latest reply: May 14, 2013 10:36 AM by SuperSchatje RSS
  • DJ GiNSU Calculating status...
    Currently Being Moderated
    Mar 14, 2013 3:15 PM (in response to tessmar)

    Tessmar -

     

    The pop-up 'server disconnected' will only show up after the problem has gotten so bad that the OS cant recover from the disconnects every five minutes.  The syslog utility shows a stream of errors long before the degradation is bad enough to cause the popup. AFP is a pretty robust protocol, which usually recovers.  If you use CIFS instead of AFP, you will find that it just outright crashes after twenty hours connected, the session just crashes instead of logging disconnects.  In either case, its a major problem.

  • tessmar Calculating status...
    Currently Being Moderated
    Mar 15, 2013 10:47 AM (in response to DJ GiNSU)

    I'm not sure that your comment about when the popup appears matches my vague perception of at least the last occurence I saw of it...which seemed to be a very isolated event. I do think that implicit in your comment is the realization that it is a timing issue-and that does match the last one of these I saw. But I have not retained the information about that single event so can only relate that I believe it was something that happened on waking the system...and was concurrent with the first occurence in the log of the disconnect.

     

    I certainly agree that the problem is more than annoying. While I have found what appears to be an ironclad workaround for my situation, had I not had a power failure last December it is unlikely I ever would have learned about the issue and, worse still, at some point my TC might have just died because of the continual partial spinups associated with the errors-and I would never have known those had happened or why the drive failed.

     

    Having said that I am skeptical that Apple will acknowledge the problem or address it in any way (but I'd love to be surprised and proven wrong). They do not seem to be geared to the way some of their more sophisticated users utilize their products. I always thought that if you stay in their world view you'd be OK, but did not realize that view was as narrow as it is.

     

    While I have 40 years in IT my technical competence at this level is not, IMO, super deep, and yet I keep stumbling into "gotchas" with relatively simple things I'm trying to do which, I believe, are going to be increasingly common- such as having multiple iTunes libraries, trying to use the iPhone as a remote control for music stored somewhere else and played through remote speakers, or, in this case, storing data on a network device for access from multiple devices and locations. In this case I think Apple really is setup for the TC to be used as a TM, period, even though that view conflicts some of their own advertising about the product.

     

    If that perception is valid, their deafening silence to date about issues documented in this thread and elsewhere will not, I believe, change.

     

    But I hope your input to Apple's executive resolution process proves me wrong.

  • DJ GiNSU Level 1 Level 1 (0 points)
    Currently Being Moderated
    Mar 15, 2013 9:36 PM (in response to tessmar)

    Look, this is very simple to duplicate, and I can almost guarantee it happens on every single time capsule running a 7.6.x firmware.

     

    1.  Open terminal on your mac, and type: sudo syslog -w

     

    2.  Open finder, locate your time capsule in the left pane, and browse to its internal drive

     

    3.  Wait 24 hours.

     

    4.  Observe this pattern showing up every 5 minutes in your syslog.

     

    Mar 15 23:10:12 zuko.local KernelEventAgent[41] <Notice>: tid 00000000 received event(s) VQ_NOTRESP (1)

    Mar 15 23:10:12 zuko.local KernelEventAgent[41] <Notice>: tid 00000000 type 'afpfs', mounted on '/Volumes/oldvault', from '//Kalani%20Thompson@Auryn._afpovertcp._tcp.local/oldvault', not responding

    Mar 15 23:10:12 zuko kernel[0] <Debug>: ASP_TCP Disconnect: triggering reconnect by bumping reconnTrigger from curr value 50 on so 0xffffff803a6d2390

    Mar 15 23:10:12 zuko kernel[0] <Debug>: ASP_TCP asp_tcp_usr_control: invalid kernelUseCount 0

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect started /Volumes/oldvault prevTrigger 50 currTrigger 51

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  doing reconnect on /Volumes/oldvault

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  posting to KEA EINPROGRESS for /Volumes/oldvault

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  Max reconnect time: 30 secs, Connect timeout: 15 secs for /Volumes/oldvault

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  connect to the server /Volumes/oldvault

    Mar 15 23:10:12 zuko.local KernelEventAgent[41] <Notice>: tid 00000000 found 1 filesystem(s) with problem(s)

    Mar 15 23:10:12 zuko.local KernelEventAgent[41] <Notice>: tid 00000000 received event(s) VQ_NOTRESP (1)

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  Logging in with uam 13 /Volumes/oldvault

    Mar 15 23:10:12 zuko kernel[0] <Debug>: AFP_VFS afpfs_DoReconnect:  get the reconnect token

     

    AFP is a fairly robust protocol and just because the OS isn't throwing a dialog on the screen to let you know it's given up, doesn't mean there's not a problem.

     

    If you would like to see a more catastrophic failure, replace step 2 with this:  Finder -> Go -> Connect to Server.  Type: cifs://[your time capsule name]/data

     

    And in 24 hours you will see a complete failure, resulting in a lock-up of the network file service.

     

    I'm already past the "let's try and get tech support to fix this" and neck deep in "It's time to talk to the execs and do something about Q/C" because the real problem is not that my time capsule doesn't do what it's supposed to.  The real problem is, they've released a firmware into the wild for the last year that doesn't work properly, and despite my arm-waving and having done all the troubleshooting and legwork for them to pass a ticket to the developers.... nobody at Apple seems to care.

     

    I'll let you know how this goes. 

  • SBeattie2 Level 1 Level 1 (40 points)
    Currently Being Moderated
    Mar 17, 2013 10:11 PM (in response to DJ GiNSU)

    If you would like to see a more catastrophic failure, replace step 2 with this:  Finder -> Go -> Connect to Server.  Type: cifs://[your time capsule name]/data

     

    I am trying this and I will report back to you what I find.  As far as I can tell there is no difference between smb://[your time capsule name]/data and cifs://[your time capsule name]/data.  After the connection to server is has been made - the Get Info option on any file/folder in the share - shows as being connected via smb://[your time capsule name]/data/filename

     

    I use smb:// quite frequently the Time Capsule (and other network drives) - and can leave the shares connected for days - without any problems.  Upon wake from sleep there will be an occasional SMB message in the log - tied with the wake from sleep event - but there is never any perceived "loss of connection" - other than a hesitation while the share reconnects (after wake) as expected.

     

    I'll check the MacBook Pro - with the share mounted via cifs:// - in 24 hours and report my findings.  The MacBook Pro is currently on AC power and set not to sleep for the purpose of this exercise.

     

    ~Scott

  • DJ GiNSU Level 1 Level 1 (0 points)
    Currently Being Moderated
    Mar 18, 2013 5:58 PM (in response to SBeattie2)

    Some good news.  Apple's executive relations connected me with a tech, who took all of the troubleshooting information down (he was very receptive) and the issue is now going to engineering.  He's going to call me tomorrow for a follow-up.  More to come.

  • SBeattie2 Level 1 Level 1 (40 points)
    Currently Being Moderated
    Mar 19, 2013 8:16 AM (in response to DJ GiNSU)

    To DJ - Glad to hear you have made contact with Apple.  You might want to reference Applecare Case# 281555173 - as that case contains all the information that I provided them when I did extensive troubleshooting with one of their senior engineers over a year ago regarding the issues mentioned in this thread.

     

    Best of luck - and please report back.

     

    ~Scott

  • SBeattie2 Level 1 Level 1 (40 points)
    Currently Being Moderated
    Mar 19, 2013 8:18 AM (in response to SBeattie2)

    I have had a share mounted on the Time Capsule for over 33 hours via cifs:// without any issue.  I am able to use the share and I do not see any messages in the log related to this share.  It appears to be working fine.

     

    ~Scott

  • tessmar Level 1 Level 1 (0 points)
    Currently Being Moderated
    Mar 19, 2013 11:39 AM (in response to DJ GiNSU)

    I think it is worth noting (and hopefully to the tech who contacts you, DJ) that there seems to be multiple situations displaying similar symptoms...i.e the disconnect/reconnect sequence, TC partial drive spinup, and, in many cases, the accompanying popup notification.

     

    When I first started lookiing into this I seem to recall multiple threads on the forum with those characteristics but what appear to be differing accompanying circumstances. But all had in common the above symptoms.

  • CMRLunatic Calculating status...
    Currently Being Moderated
    Mar 20, 2013 1:52 PM (in response to tessmar)

    Adding to this a bit...

     

    I have a USB Drive plugged into an airport router, being used by iTunes.  I get the infamous popup in the mornings.

     

    Last night I purchased the season pass for The Following.   I have iTunes set to NOT download multiple files at once.  So, in the morning it was still downloading - and there wasn't any popup.

     

    Activity all night - no popup.     No activity - popup.

  • DJ GiNSU Level 1 Level 1 (0 points)
    Currently Being Moderated
    Mar 28, 2013 7:18 PM (in response to CMRLunatic)

    I have great news.   I've been working with Apple for a week now, performing log collections and exchanging data and tech I've been working with says engineering has been able to duplicate the problem, and is working on a fix. 

     

    They say they do not have an ETA as of yet, but that the fix will probably be deployed as a firmware or software update. 

     

    This means, hopefully, many of you who are still having issues, will have your time capsule and AE file sharing problems resolved in the near future.

     

    Apple has been extremely receptive, and I could not ask for better treatment.

  • SBeattie2 Level 1 Level 1 (40 points)
    Currently Being Moderated
    Mar 28, 2013 9:14 PM (in response to DJ GiNSU)

    To DJ - Thank you for following through with this and spending your valuable time helping Apple reproduce the problem.  Hopefully they will be able to fix it quickly and this thread will finally come to an end.

     

    With so many people having slightly different occurences of the problem - did Apple happen to give any indication of what the exact cause of the problem might be?

     

    Please keep us posted.

     

    ~Scott

  • tessmar Level 1 Level 1 (0 points)
    Currently Being Moderated
    Mar 28, 2013 9:19 PM (in response to DJ GiNSU)

    DJ

    Thanks for your persistence. I was not optimistic about Apple's ability to reproduce the errors. And absent that it seemed the discussions were going nowhere. It is a credit to the company that they do have an executive resolution process...and somehow almost a shame that this is the card that had to be played in order to make progress...but kudos to you for the effort!

  • Menge Calculating status...
    Currently Being Moderated
    Apr 5, 2013 5:01 AM (in response to DJ GiNSU)

    Thanks for finally getting Apple to see this issue! I'm also a victim of it and have always kept my AirPort Extreme on firmware 7.5.2 to avoid this.

     

    I would love to see this fixed so I can keep my router up-to-date. Please let us know if you have any updates on this issue!

  • C.Jason Level 1 Level 1 (0 points)
    Currently Being Moderated
    Apr 5, 2013 7:36 AM (in response to Folbo)

    Hi everyone, I've been following this discussion for a long while and have seen how it's evolved in the past year or so. Like all of you, it started for me when I kept getting harmless "Server connections interrupted" pop ups in a seemingly distinct pattern / condition.

     

    I don't use time machine to back up my files, instead I use it purely as a wireless external HD. I added the "Data" volume into my login item list and that's how I've been using it since the original problem surfaced after an earlier hardware update.

     

    But just when I was starting to get used to the annoyance, it stopped completely one day (for a while, I thought the nightmare was finally over, but it didn't). I'm not sure but I have a feeling that the behavior change following a recent hardware update, 7.6.3?

     

    My "Data" volume stays connected for a couple of minutes after booting up, but then it starts to unmount if there is no read or write activity after that first few minutes. I will have to click on finder, select my TC under the list of shared devices, double click on Data to have the disk mounted back.

     

    But this time round, it doesn't stay mounted longer than a minute. I have my entire itunes music library on the TC disk, and the problem is so ridiculously bad that everytime i pause a song halfway through playback, pause a movie on quicktime player half way through playback, the disk disconnects!

     

    My only work around is to have one mp3 on the TC disk to play on quicktime, on loop, and at no volume.

     

    Anyone having the same problem?

  • Michael Wakefield1 Level 1 Level 1 (5 points)
    Currently Being Moderated
    Apr 6, 2013 8:30 AM (in response to DJ GiNSU)

    Looking forward to the fix, thanks for the persistence, DJ!

     

    Have just reverted back to 7.5.2 myself to try to stop the endless stream of afp popups. So far so good. I had hoped that the recent upgrade to 7.6.3. would be the fix, but it seemed to make things even worse. Am using attached USB drive for TC as well as general NAS, everything has been OK while reading/writing to the volume, but if the connection has been idle for more than a few seconds, the afp drops start showing up in the log.

Actions

More Like This

  • Retrieving data ...

Bookmarked By (22)

Legend

  • This solved my question - 10 points
  • This helped me - 5 points
This site contains user submitted content, comments and opinions and is for informational purposes only. Apple disclaims any and all liability for the acts, omissions and conduct of any third parties in connection with or related to your use of the site. All postings and use of the content on this site are subject to the Apple Support Communities Terms of Use.