1.2.0 issues



  • I also upgraded to 1.2.0, feedback:

    1. I also have the problem with host unreachable
      If I telnet the host from outside on 9982 its ok, I am also getting new contracts

    2. I was hosting around 400-500GB of files, 2 shares of 500GB stored on a dedicated USB hdd just for SIA.
      It took around 3 days to upgrade!!!

    3. I would like to see some basic stats about how much bandwith (total and live) SIA is using.



  • @Taek said in 1.2.0 issues:

    I found out aprox. 140TB per path is maximum amount of space possible to assign

    I don't remember coding anything like that. Are you sure it's not a limitation of your local filesystem?

    In following example, I was trying to add 200TB storage.
    File system is ReFS, similar behavior observed when using NTFS.

    In previous version of Sia, 1PB storage was not an issue.

    Please see enlcosed140TB limit


    @lordali said in 1.2.0 issues:

    1. I was hosting around 400-500GB of files, 2 shares of 500GB stored on a dedicated USB hdd just for SIA.
      It took around 3 days to upgrade!!!

    2. I would like to see some basic stats about how much bandwith (total and live) SIA is using.

    I do concur it does take long time to upgrade. It become an issue when shared folders are hundreds of Gigabytes and nightmare when Terabytes or more.


    Another issue is time stamps on wallet transaction logs are wrong. See enclosed: enclosed



  • Would be nice to see some feedback from @Core-Dev regarding mentioned issues, sometime this year preferably.


  • Global Moderator

    @uzej not sure if you realize, but @Taek is a Core Dev



  • @zherbert said in 1.2.0 issues:

    @uzej not sure if you realize, but @Taek is a Core Dev

    Whom I was hoping to express his thoughts on this matter.


  • admins

    Hey, sorry for the delay. I was hoping that v1.2.1 would be out on Friday and I could comment and tell you that we had fixed most of the issues. List of changes:

    1. max storage folder size increased from 256TiB to 32 PiB. I realize you are probably using a cloud drive as your storage folder, the system had originally been designed to work with hard drives. You can't do better than 32 PiB unfortunately, the disk layout only allows you to have 2^32 sectors of 4 MiB each.

    2. Multi-day upgrades should be fixed. It's a one time thing when moving from <v1.2.0 to >v1.2.0, but if you haven't made that leap yet it'll be a lot faster in v1.2.1. Upgrading from v1.2.0 to v1.2.1 should take a few seconds. (v1.1.2 to v1.2.1 will still take hours, but at least not days).

    3. not sure what the issue is with the wallet transaction logs.

    Would be nice to see some feedback from @Core-Dev regarding mentioned issues, sometime this year preferably.

    There's no need for that. I slept an average of 4 hours a day last week working with users and fixing bugs in the release. Some forgiveness on the response times would be appreciated.



  • @Taek said in 1.2.0 issues:

    1. not sure what the issue is with the wallet transaction logs.

    If you look at screenshot the issue is:
    transaction for -1.261 was made 17-04-19 22:34
    transaction for -2.869 was made 17-04-20 14:49

    but according the log it is look like transaction for -1.261 was made 17-04-20 which is not true.

    TL:DR if log sorting transaction 'latest at the top', then -1.261 should be above -2.869 based on time of transaction, and it is not due to the fact it has physically was a day before.

    I hope I did express myself clearly.

    @Taek said in 1.2.0 issues:

    I realize you are probably using a cloud drive as your storage folder, the system had originally been designed to work with hard drives.

    Actually several PT of storage on FreeNAS under GlusterFS shared as network storage.


  • admins

    @uzej said in 1.2.0 issues:

    @Taek said in 1.2.0 issues:

    1. not sure what the issue is with the wallet transaction logs.

    If you look at screenshot the issue is:
    transaction for -1.261 was made 17-04-19 22:34
    transaction for -2.869 was made 17-04-20 14:49

    but according the log it is look like transaction for -1.261 was made 17-04-20 which is not true.

    TL:DR if log sorting transaction 'latest at the top', then -1.261 should be above -2.869 based on time of transaction, and it is not due to the fact it has physically was a day before.

    I hope I did express myself clearly.

    Ah yes, I see the issue now. The sorting is weird, or the timestamps are weird, or something else is going on. Thanks for pointing that out, we'll investigate it further.

    @Taek said in 1.2.0 issues:
    Actually several PT of storage on FreeNAS under GlusterFS shared as network storage.

    Oh that is really interesting. Can I ask a few more questions then, I'm not really familiar with these types of setups.

    1. You have access to the physical drives, right?
    2. Does GlusterFS handle things like drive parity, or when a disk fails do you lose data?
    3. How big do these systems reasonably get?
    4. What is your primary motivation for being a host on the Sia network?

    Thanks :)



  • @Taek said in 1.2.0 issues:

    1. You have access to the physical drives, right?
    2. Does GlusterFS handle things like drive parity, or when a disk fails do you lose data?
    3. How big do these systems reasonably get?
    4. What is your primary motivation for being a host on the Sia network?
    1. To the servers that are installed in-house yes. Not in other locations.
    2. ClusterFS can operate on different levels.
    • distributed (similar to RAID0)
    • replicated (similar to RAID1)
    • distributed replicated (similar to RAID10)
    • striped (pieces of data are spread across different locations, for load reducing)
    • distributed striped (same as above but also with redundancy)
      I am using raidz3 underneath Gluster, so redundancy is taken care of on ZFS level mostly.
    1. As big as you budget can be stretched.
    2. Mostly curiosity of different approach to data handling and management. I do not think about profits at any stage yet. Maybe in year time when project mature.

    Also side note. After upgrading to 1.2.1 "host unreachable" has been fixed. It is "online" now.**



  • Hello
    Same issue as "uzej" , bu twith 1.2.2 version
    Datas from 500GB to 140 after migration
    And always "Connectability Status: Nobody is connecting to host" ...


Log in to reply