15
Oct
14

Recovery Update

(from irc, oct 15)

< leTwitch> The news is, there really isn’t much new to tell. Three different organizations spread geographically all over Texas are all struggling with a broken raid volume. We get two steps fwd and one back; be patient. There really isn’t a good way to give an eta on work like this; once we can certify that we have the volume back in a readable form, then we can start pinning down certain tasks and registering progress. Even then, ETAs will be just that *estimated* time to arrival.
< leTwitch>  The real showstopper at this juncture is that the data recovery folks can see the data, but when they ship it to the datacenter and we get it hooked up to the server, we cant. So right now what’s happening is a straight up dump of the paths and directories to a big volume (single partition). The last thing that didn’t work for us was a set of partition copies. The thing we’re doing now is the last trick up the sleeve of the engineer doing the data recovery; if this doesn’t work, we’ll be left with nothing more to do than rebuild the server and start rebuilding from archives.
My understanding is, we dont pay anything if they dont get our stuff back, so if thats how it winds up, we’ll be talking with them about getting back what we have already paid them.
< leTwitch> so if nothing else, the end of the waiting will soon come to pass.
< leTwitch> we’ll be moving forward one way or another.
< leTwitch> Heh, I guess that is news of a sort.
< leTwitch> o/

01
Oct
14

OSgrid Info

Greetings,

We have new information on the recovery process. The array was not able to be rebuilt from the cloned drives due to logical partition errors so the recovery service has performed a full recovery, restoring everything to a single 6 tb drive that the engineer said should boot and that all the files are intact. The additional costs, less the $2302.56 we have already paid them, comes to $2510. We currently have $3432.88 which would make our budget very tight (we still have hosting costs, etc).

Thank you very much for your patience and support!

12
Sep
14

Recovery details

Greetings,

The recovery firm has responded with some news, some bad, some good. The initial drive assessment determined that they found sector damage, partition corruption, filesystem corruption, and logical alignment failure. They provided an estimate for file recovery that probably exceeds our means at $4600 and would still require a lot of work to put back into place.

However during subsequent discussions with the engineer it has been suggested that cloning the drives would probably be a more viable solution. The good news is that cloning is less expensive and we would have a working partition, which would help get the grid back online sooner. They will need 4 new drives to accomplish this, and we have contacted our data center about this. We will be able to provide more detailed information, as soon as we receive it.

Thanks again for your patience and understanding.

OSgrid, Inc.

20
Aug
14

Current status

Greetings.

This post is to notify users and other concerned community that OSgrid has suffered a critical failure in one of the primary drives in our RAID10 array. This is a matter of the gravest technical concern, and while we are working diligently with our datacenter services provider to resolve this issue without data loss, it is imperative that utmost care be directed at this process. This means that it will take time to resolve; even time and the greatest of care are not guarantees that we will come through it without data loss.

Consequently, the grid is offline at present, and will be until we can get this issue resolved. We will do our best to keep everyone up to date on the issue and its resolution as we go forward, but please understand that no news is good news, and that we will have the recovery operation as our first priority until some disposition of the issue has been achieved.

We greatly appreciate your patience and understanding in this matter, and apologize for the jeopardy in which this places all of our content.

Thank you,

OSgrid, Inc.

 

18
Aug
14

Unscheduled Maintenance

Aug 18, 2014 Grid is OFFLINE for unscheduled maintenance. Follow us on twitter for updates: https://twitter.com/osgrid

Sorry for the inconvenience, we hope to have this resolved as soon as possible.

03
Mar
13

Update – Release changes for OSgrid opensimulator.0.7.6.dev.e70c71a

Hello everyone,

I have posted a new version of OSgrid Simulator release on the website today, there are some important changes that you should be aware of before upgrading to make sure you retain 100% functionality.

The main changes are in two files, first is the OpenSim.ini . The following settings need to be moved from [Startup] into their own sections.  (link for the fully updated OpenSim.ini )

OpenSim.ini changes:

_________________________

[AccessControl]
    ;# {AllowedClients} {} {Bar (|) separated list of allowed clients} {}
    ;; Bar (|) separated list of viewers which may gain access to the regions.
    ;; One can use a substring of the viewer name to enable only certain
    ;; - "Imprudence" has access
    ;; - "Imprudence 1.3" has access
    ;; - "Imprudence 1.3.1" has no access
    ; AllowedClients =

    ;# {BannedClients} {} {Bar (|) separated list of banned clients} {}
    ;# Bar (|) separated list of viewers which may not gain access to the regions.
    ;; - "Imprudence" has no access
    ;; - "Imprudence 1.3" has no access
    ;; - "Imprudence 1.3.1" has access
    ; BannedClients =
[Permissions]
    ;# {permissionmodules} {} {Permission modules to use (may specify multiple modules, separated by comma} {} DefaultPermissionsModule
    ;; Permission modules to use, separated by comma.
    ;; Possible modules are DefaultPermissionsModule, PrimLimitsModule
    permissionmodules = DefaultPermissionsModule

    ;# {serverside_object_permissions} {permissionmodules:DefaultPermissionsModule} {Activate permission handling by the sim?} {true false} true
    ;; These are the parameters for the default permissions module
    ;;
    ;; If set to false, then, in theory, the server never carries out
    ;; permission checks (allowing anybody to copy
    ;; any item, etc.  This may not yet be implemented uniformally.
    ;; If set to true, then all permissions checks are carried out
    serverside_object_permissions = true

    ;# {allow_grid_gods} {} {Allow grid gods?} {true false} false
    ;; This allows users with a UserLevel of 200 or more to assume god
    ;; powers in the regions in this simulator.
    allow_grid_gods = false

    ;; This allows some control over permissions
    ;; please note that this still doesn't duplicate SL, and is not intended to
    ;# {region_owner_is_god} {} {Allow region owner gods} {true false} true
    ;; Allow region owners to assume god powers in their regions
    region_owner_is_god = true

    ;# {region_manager_is_god} {} {Allow region manager gods} {true false} false
    ;; Allow region managers to assume god powers in regions they manage
    ; region_manager_is_god = false

    ;# {parcel_owner_is_god} {} {Allow parcel owner gods} {true false} true
    ;; Allow parcel owners to assume god powers in their parcels
    ; parcel_owner_is_god = true

    ;# {simple_build_permissions} {} {Allow building in parcel by access list (no groups)} {true false} false
    ;; More control over permissions
    ;; This is definitely not SL!
    ;; Provides a simple control for land owners to give build rights to
    ;; specific avatars in publicly accessible parcels that disallow object
    ;; creation in general.
    ;; Owners specific avatars by adding them to the Access List of the parcel
    ;; without having to use the Groups feature
    ; simple_build_permissions = false
[Map]
    ;# {GenerateMaptiles} {} {Generate map tiles?} {true false} true
    ;; Map tile options. You can choose to generate normal maptiles or nominate an uploaded texture to
    ;; be the map tile using the MaptileStaticUUID parameter in this section or for individual regions in
    ;; the regions config file(s).  If you do not want to upload map tiles at all, then you will need
    ;; to disable the MapImageServiceModule entirely.
    GenerateMaptiles = true

    ;# {MaptileRefresh} {GenerateMaptiles} {Maptile refresh period?} {} 0
    ;; If desired, a running region can update the map tiles periodically
    ;; to reflect building activity. This names no sense of you don't have
    ;; prims on maptiles. Value is in seconds.
    ; MaptileRefresh = 0

    ;# {MaptileStaticUUID} {} {Asset ID for static map texture} {} 00000000-0000-0000-0000-000000000000
    ;; If not generating maptiles, use this static texture asset ID
    ; MaptileStaticUUID = "00000000-0000-0000-0000-000000000000"

    ;# {TextureOnMapTile} {} {Use terrain textures for map tiles?} {true false} true
    ;; Use terrain texture for maptiles if true, use shaded green if false
    ; TextureOnMapTile = true

    ;# {DrawPrimOnMapTile} {} {Draw prim shapes on map tiles?} {true false} false
    ;; Draw objects on maptile.  This step might take a long time if you've
    ;; got a large number of objects, so you can turn it off here if you'd like.
    ; DrawPrimOnMapTile = true

_________________________

The following settings need to be moved from the [Startup] in OpenSim.ini to a new section of the GridCommon.ini. (link for fully updated GridCommon.ini)

GridCommon.ini changes

_________________________

[Hypergrid]
    ; Uncomment the variables in this section only if you are in
    ; Hypergrid configuration. Otherwise, ignore.

    ;# {HomeURI} {Hypergrid} {The Home URL of this world} {}
    ;; If this is a standalone world, this is the address of this instance.
    ;; If this is a grided simulator, this is the address of the external robust server that
    ;; runs the UserAgentsService.
    ;; For example http://myworld.com:9000 or http://myworld.com:8002
    ;; This is a default that can be overwritten in some sections.
    HomeURI = "http://hg.osgrid.org:80/user"

    ;# {GatekeeperURI} {Hypergrid} {The URL of the gatekeeper of this world} {}
    ;; If this is a standalone world, this is the address of this instance.
    ;; If this is a grided simulator, this is the address of the external robust server
    ;; that runs the Gatekeeper service.
    ;; For example http://myworld.com:9000 or http://myworld.com:8002
    ;; This is a default that can be overwritten in some sections.
    GatekeeperURI = "http://hg.osgrid.org:80"

_________________________

If you have any questions or need help with this update please join us on our IRC chat channel at irc.freenode.net #osgrid or you can use the web chat or on the OSgrid forums here: http://forums.osgrid.org/viewtopic.php?f=24&t=4708

Good luck everyone,

Dan

14
Dec
12

Grid Downtime scheduled for 12/17/2012 (Monday)

Hello Everyone,

As some of you may or may not have heard our hosting agreement with University of California has come to an end, the grant that was funding the project that hosted our server has ended and will not be renewed. What does this mean for OSgrid? It means we will be moving the server this coming monday to a new hosting provider, on Monday 12/17/2012 around 9am EST we will be taking the grid and websites down for a short bit while we transition the hardware to the new host, we are not quite sure how long the transition will take exactly so we do not want to give any estimates at this time, but if all goes well we should be back up sometime later that day, however we need to be prepared for the unexpected. We will be giving updates throughout the day marking progress as we near completion. We highly recommend that you log out of the grid during this time and especially if you do manage to stay inside the grid that you do not do anything with your inventory or do any building in your simulator as it will likely result in lost data and possible corruption of your builds. As part of this transition to cut down on costs shortly after this server move we will also be consolidating 2 of our plaza servers into 1 new more powerful server, when all is said and done our monthly costs will be about the same as they were before this. I would like to thank everyone who has donated their time and monetary donations as well, without this we can not survive, if you are not already donating to the grid you can find out more about how to here : http://www.osgrid.org/index.php/donate, thank you everyone if you have any questions you can visit us at our IRC/Web Chat here : http://www.osgrid.org/index.php/freenode or via your favorite IRC client on irc.freenode.net #osgrid. If you want to follow progress throughout the day during the downtime, you can follow us on twitter http://twitter.com/osgrid

Michael Emory Cerquoni
(Nebadon Izumi)




Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 193 other followers

Copyright © 2007-2010 OSGrid, Inc. - All rights reserved, except where noted.

The OSgrid Logo, and the word 'OSgrid' are trademarks of OSGrid, Inc. Usage of these terms elsewhere is allowed under certain conditions.


Follow

Get every new post delivered to your Inbox.

Join 193 other followers