Quantcast
Channel: StudioSysAdmins Message Board
Viewing all 3749 articles
Browse latest View live

Top 2 at Aspera gone

0
0
Top 2 at Aspera gone
posted by Ken Spickler on June 7, 2017, 1:45 p.m.
Surely I'm not the only one who's questioning the future of the products in their absence.

http://www.sportsvideo.org/2017/06/06/aspera-founders-michelle-munson-serban-simu-leave-ibm/

Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Surely I'm not the only one who's questioning the future of the products in their absence.

http://www.sportsvideo.org/2017/06/06/aspera-founders-michelle-munson-serban-simu-leave-ibm/


Monitoring Sentinel dongle / license server

0
0
Monitoring Sentinel dongle / license server
posted by Jean-Francois Panisset on June 7, 2017, 7:05 p.m.
The Sentinel Protection Server presents a web interface on port 6002, but that uses a Java app which gets downloaded into the browser, which means it might as well require a PhD in astrophysics to get and keep working. And it doesn't allow you to integrate it into a monitoring infrastructure.

Wireshark showed that the Java app was just making http requests itself, but somehow I wasn't able to replicate the correct sequence with wget.

Then found this guy's script:

https://wiki.xymonton.org/doku.php/monitors:sentinel

Turns out you first have to GET a HTML document passing the parameters of the license / dongle you want to get, then quickly GET a XML document with the results. Apparently "http is stateless" doesn't apply in this case.

Thank you script author Richard Finegold for figuring this out.

JF



Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
The Sentinel Protection Server presents a web interface on port 6002, but that uses a Java app which gets downloaded into the browser, which means it might as well require a PhD in astrophysics to get and keep working. And it doesn't allow you to integrate it into a monitoring infrastructure.

Wireshark showed that the Java app was just making http requests itself, but somehow I wasn't able to replicate the correct sequence with wget.

Then found this guy's script:

https://wiki.xymonton.org/doku.php/monitors:sentinel

Turns out you first have to GET a HTML document passing the parameters of the license / dongle you want to get, then quickly GET a XML document with the results. Apparently "http is stateless" doesn't apply in this case.

Thank you script author Richard Finegold for figuring this out.

JF



FREE webinar

0
0
FREE webinar
posted by Gene Leyzarovich on June 8, 2017, 11:05 a.m.
http://www.broadcastbeat.com/jetstor-raidix-invite-you-to-a-webinar-cut-your-storage-costs-by-70-percent/ To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
http://www.broadcastbeat.com/jetstor-raidix-invite-you-to-a-webinar-cut-your-storage-costs-by-70-percent/ To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Create Comp issue (hanging)

0
0
Create Comp issue (hanging)
posted by Andrew Lim on June 8, 2017, 3:22 p.m.

Hi all,

I've recently started having an issue creating comps in Hiero/Nuke Studio where the progress bar hangs at 0% and the app becomes unresponsive i.e. I have to Force Quit. The frustrating thing is that it generates the required folders on my disk and creates the Nuke script but it doesn't make the track on the timeline. I'm using 10.5v4 but have the same behaviour on v2 and 10.0.5 as well.

Foundry support have been close to useless saying "thanks we recreated it, case is closed, bye!" (literally). For those interested here is that case...

https://support.foundry.com/hc/en-us/articles/115000755284-Nuke-Create-Comp-Special-export-panel-does-not-reflect-actual-progress-

 

I'm on Mac OS but they recreated it on Windows however this was not a problem til I updated to macos sierra.

Any sort of clue or suggestion would be appreciated.

Thanks

Andrew

Thread Tags:
  hiero, studio, nuke, foundry 

0 Responses   0 Plus One's   0 Comments  
 

Hi all,

I've recently started having an issue creating comps in Hiero/Nuke Studio where the progress bar hangs at 0% and the app becomes unresponsive i.e. I have to Force Quit. The frustrating thing is that it generates the required folders on my disk and creates the Nuke script but it doesn't make the track on the timeline. I'm using 10.5v4 but have the same behaviour on v2 and 10.0.5 as well.

Foundry support have been close to useless saying "thanks we recreated it, case is closed, bye!" (literally). For those interested here is that case...

https://support.foundry.com/hc/en-us/articles/115000755284-Nuke-Create-Comp-Special-export-panel-does-not-reflect-actual-progress-

 

I'm on Mac OS but they recreated it on Windows however this was not a problem til I updated to macos sierra.

Any sort of clue or suggestion would be appreciated.

Thanks

Andrew

Production Recruitment Open Day

0
0
Production Recruitment Open Day
posted by Jill Wallace on June 9, 2017, 6:48 a.m.

Axis Studios Glasgow have announced an exciting opportunity for local production talent to participate in a Production Open Day on Saturday 24th June 2017. Axis welcome both experienced industry professionals; Producers, Production Managers, Production Coordinators and Assistants, and also participants from outside the media industry, who are interested in working with our Emmy Award winning team.

  • Do you like to take responsibility for ensuring the problem is solved?
  • Are you able to anticipate a problem and either prevent it or limit the damage?
  • Can you see past the noise?
  • Are you able to delegate responsibility effectively?
  • Do you respond pro-actively to stress?

If the answer is Yes! Axis want to hear from you.

There are two slots available 10.00 - 13.30 and 14.30 – 18.00 and will consist of both individual and group activities in our Glasgow studio.

Spaces are limited so send your CV and cover letter over to openday@axisanimation.com for further information and to secure your place.

Closing date for entry is Sunday 18th June.

Thread Tags:
  production, jobs, Axis, Animation 

0 Responses   0 Plus One's   0 Comments  
 

Axis Studios Glasgow have announced an exciting opportunity for local production talent to participate in a Production Open Day on Saturday 24th June 2017. Axis welcome both experienced industry professionals; Producers, Production Managers, Production Coordinators and Assistants, and also participants from outside the media industry, who are interested in working with our Emmy Award winning team.

  • Do you like to take responsibility for ensuring the problem is solved?
  • Are you able to anticipate a problem and either prevent it or limit the damage?
  • Can you see past the noise?
  • Are you able to delegate responsibility effectively?
  • Do you respond pro-actively to stress?

If the answer is Yes! Axis want to hear from you.

There are two slots available 10.00 - 13.30 and 14.30 – 18.00 and will consist of both individual and group activities in our Glasgow studio.

Spaces are limited so send your CV and cover letter over to openday@axisanimation.com for further information and to secure your place.

Closing date for entry is Sunday 18th June.

Any facilities using pcoip at greater than 50ms latency?

0
0
Any facilities using pcoip at greater than 50ms latency?
posted by TimLeDoux on June 9, 2017, 5:02 p.m. (1 day ago)

From testing it appears that this is viable if you are using a mouse, but because many artists use tablets it is not an option for an entire facility to make that transition.  Wacoms at this latency are too laggy and visible acceleration with zero clients and host cards using "local cursor" is faulty (at least in windows).  

First post! Hope this is in the right place.  Thanks.

Thread Tags:
  teradici 

0 Responses   0 Plus One's   0 Comments  
 

From testing it appears that this is viable if you are using a mouse, but because many artists use tablets it is not an option for an entire facility to make that transition.  Wacoms at this latency are too laggy and visible acceleration with zero clients and host cards using "local cursor" is faulty (at least in windows).  

First post! Hope this is in the right place.  Thanks.

Purchasing City Geometry data

0
0
Purchasing City Geometry data
posted by Todd Smith on June 16, 2017, 3:10 p.m. (1 day ago)
Hi All,

We are looking to purchase geometry data for various places around the globe.  Does anyone have a good vendor they use for this?

Thanks,

Todd Smith
Head of Information Technology

soho vfx 
40 Hanna Ave. Suite 403, Toronto, Ontario M6K 0C3
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Hi All,

We are looking to purchase geometry data for various places around the globe.  Does anyone have a good vendor they use for this?

Thanks,

Todd Smith
Head of Information Technology

soho vfx 
40 Hanna Ave. Suite 403, Toronto, Ontario M6K 0C3

MacDevOps:YVR 2017 videos

0
0
MacDevOps:YVR 2017 videos
posted by Mathieu Mauser on June 21, 2017, 4:25 p.m.
FYI

For those who couldnt make it this year or were there and want to watch the videos. We put them on YouTube this year.


Lots of good stuff. Talks from Google, Facebook, Dropbox, Airbnb, Square, Disney Animation, and many more.

Cheers,

:) 

Mat X

----------------





Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
FYI

For those who couldnt make it this year or were there and want to watch the videos. We put them on YouTube this year.


Lots of good stuff. Talks from Google, Facebook, Dropbox, Airbnb, Square, Disney Animation, and many more.

Cheers,

:) 

Mat X

----------------






Patch Management Policy

0
0
Patch Management Policy
posted by Ryan Lynch on June 23, 2017, 5:20 p.m. (1 day ago)
Does anyone have a good patch management policy that they would like to share? I am creating one to satisfy MPAA and studio audits and would like to compare to what other folks are using.

Thanks!

Ryan
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Does anyone have a good patch management policy that they would like to share? I am creating one to satisfy MPAA and studio audits and would like to compare to what other folks are using.

Thanks!

Ryan

Broken hyper threading on skylake/kabylake processors

0
0
Broken hyper threading on skylake/kabylake processors
posted by Matt Plec on June 26, 2017, 1:30 p.m. (1 day ago)
Got this today from our lead unix sysadmin, a bit out of my area but looks relevant to many here:

https://lists.debian.org/debian-devel/2017/06/msg00308.html
This warning advisory is relevant for users of systems with the Intel
processors code-named "Skylake" and "Kaby Lake". These are: the 6th and
7th generation Intel Core processors (desktop, embedded, mobile and
HEDT), their related server processors (such as Xeon v5 and Xeon v6), as
well as select Intel Pentium processor models.
TL;DR: unfixed Skylake and Kaby Lake processors could, in some
situations, dangerously misbehave when hyper-threading is enabled.
Disable hyper-threading immediately in BIOS/UEFI to work around the
problem. Read this advisory for instructions about an Intel-provided
fix.

Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Got this today from our lead unix sysadmin, a bit out of my area but looks relevant to many here:

https://lists.debian.org/debian-devel/2017/06/msg00308.html
This warning advisory is relevant for users of systems with the Intel
processors code-named "Skylake" and "Kaby Lake". These are: the 6th and
7th generation Intel Core processors (desktop, embedded, mobile and
HEDT), their related server processors (such as Xeon v5 and Xeon v6), as
well as select Intel Pentium processor models.
TL;DR: unfixed Skylake and Kaby Lake processors could, in some
situations, dangerously misbehave when hyper-threading is enabled.
Disable hyper-threading immediately in BIOS/UEFI to work around the
problem. Read this advisory for instructions about an Intel-provided
fix.

Flame 2018 FlexLM tokens now documented

0
0
Flame 2018 FlexLM tokens now documented
posted by Jean-Francois Panisset on June 29, 2017, 1:40 p.m.
If you too have a FlexLM / AdLM monitoring infrastructure and have been trying to make sense of the Flame 2018 tokens, these are now documented at: https://knowledge.autodesk.com/customer-service/network-license-administration/managing-network-licenses/interpreting-your-license-file/feature-codes/2018-flexnet-feature-codes JF To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
If you too have a FlexLM / AdLM monitoring infrastructure and have been trying to make sense of the Flame 2018 tokens, these are now documented at: https://knowledge.autodesk.com/customer-service/network-license-administration/managing-network-licenses/interpreting-your-license-file/feature-codes/2018-flexnet-feature-codes JF To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Magical Sym-link-ery

0
0
Magical Sym-link-ery
posted by Trevor Rowland on July 6, 2017, 3:19 p.m.

Hi all,
I have a question about actual symlink cost, both on server load and actual resolution times (both sides of same coin arguably)
You may see from my linkedin profile I am already a level 11 wizard, however I have just stopped to truly think about the (inherited) current system at work, and the more I think of it, the more it feels like a dr Who episode, where - without the magical "-alh" after your "ls" - you completely forget the terrible evil exists.... and it just sits there growing....

Our user directories (centos6.5) are mount point symlinks "/mnt/symlink_to_user_directory" to the main drive that hosts them - basically meaning the host drive can be moved at any point with zero downtime.
....so to open a terminal "<user>@<host>:" we have already jumped through one symlink, but (and here is where I'm looking for a level 20 wizard to jump in and tell me Im wrong) that link is resolved at this point, so it doesnt count.
Then we try to run an app - lets call it "DooHini". We know about DooHini because of the .cshrc file that adds the symlink to the scripts directory to our environment. Wait... I just said "symlink" again. We set a couple of environment variables (including symlink containing paths to LD_LIBRARY_PATH) then follow the symlink to the DooHini application. (Again - level 20 wizards all links are resolved in the execution context of the app and do not count at this point toward our maximm symlink depth - right???)
We then open the published alembic file which is ...you guessed it... a symlink to the file that is hosted on a symlinked drive array (meaning the host drive can be moved at any point with zero downtime....)

Strangely we never hit the maximum symlink depth, but out servers are curiously slow...
I need actual real world data to be able to prove to the powers that be this needs attention.

So for all you level 20 wizards out there:
- Does anyone know - for definate - the actual server side cost of resolving a symlink?
- Does anyone know the cost of adding a symlinked library to the LD_LIBRARY_PATH for an application ?
- Does anyone know the cost of adding a symlink to the PYTHONPATH?
- Are there any tools or things that can let me know "to get to point X in this execution context you have resolved Y symlinks"
- Is this all just a part of the passage to level 12 wizardy ? Should I just nod, smile and think about something else ?


And for everyone else:
- how many symlinks do you traverse per execution context ?
- Are you sure about your above answer ?

 

Sorry for the really long post, and if this is in the wrong spot.

Thanks for any pointers!

 

Trevor

Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 

Hi all,
I have a question about actual symlink cost, both on server load and actual resolution times (both sides of same coin arguably)
You may see from my linkedin profile I am already a level 11 wizard, however I have just stopped to truly think about the (inherited) current system at work, and the more I think of it, the more it feels like a dr Who episode, where - without the magical "-alh" after your "ls" - you completely forget the terrible evil exists.... and it just sits there growing....

Our user directories (centos6.5) are mount point symlinks "/mnt/symlink_to_user_directory" to the main drive that hosts them - basically meaning the host drive can be moved at any point with zero downtime.
....so to open a terminal "<user>@<host>:" we have already jumped through one symlink, but (and here is where I'm looking for a level 20 wizard to jump in and tell me Im wrong) that link is resolved at this point, so it doesnt count.
Then we try to run an app - lets call it "DooHini". We know about DooHini because of the .cshrc file that adds the symlink to the scripts directory to our environment. Wait... I just said "symlink" again. We set a couple of environment variables (including symlink containing paths to LD_LIBRARY_PATH) then follow the symlink to the DooHini application. (Again - level 20 wizards all links are resolved in the execution context of the app and do not count at this point toward our maximm symlink depth - right???)
We then open the published alembic file which is ...you guessed it... a symlink to the file that is hosted on a symlinked drive array (meaning the host drive can be moved at any point with zero downtime....)

Strangely we never hit the maximum symlink depth, but out servers are curiously slow...
I need actual real world data to be able to prove to the powers that be this needs attention.

So for all you level 20 wizards out there:
- Does anyone know - for definate - the actual server side cost of resolving a symlink?
- Does anyone know the cost of adding a symlinked library to the LD_LIBRARY_PATH for an application ?
- Does anyone know the cost of adding a symlink to the PYTHONPATH?
- Are there any tools or things that can let me know "to get to point X in this execution context you have resolved Y symlinks"
- Is this all just a part of the passage to level 12 wizardy ? Should I just nod, smile and think about something else ?


And for everyone else:
- how many symlinks do you traverse per execution context ?
- Are you sure about your above answer ?

 

Sorry for the really long post, and if this is in the wrong spot.

Thanks for any pointers!

 

Trevor

Isilon Continuous Availability bug + Disabling SMB3 on Isilon

0
0
Isilon Continuous Availability bug + Disabling SMB3 on Isilon
posted by Julian Firminger on July 11, 2017, 5:55 a.m. (2 days ago)
Isilon-peeps,

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server. - I will post more on this if/when we know it.

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS. Persisting sessions will continue to be multichannel until dropped / re-established.

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node). Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness. This will effectively leave CA turned on (on enabled shares) but prevent clients from using it. This will force reset any persisting SMB3-CA sessions.

To disable SMB3 entirely:
# isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from. Persisting sessions will continue to be SMB3 until dropped / re-established.

The last two are USE AT OWN PERIL. And I have been advised that these changes should be done by an EMC engineer. Especially the lwwit stop. This can have drastic and immediate effects across all SMB sessions.


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Isilon-peeps,

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server. - I will post more on this if/when we know it.

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS. Persisting sessions will continue to be multichannel until dropped / re-established.

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node). Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness. This will effectively leave CA turned on (on enabled shares) but prevent clients from using it. This will force reset any persisting SMB3-CA sessions.

To disable SMB3 entirely:
# isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from. Persisting sessions will continue to be SMB3 until dropped / re-established.

The last two are USE AT OWN PERIL. And I have been advised that these changes should be done by an EMC engineer. Especially the lwwit stop. This can have drastic and immediate effects across all SMB sessions.


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands

TRe: Isilon Continuous Availability bug + Disabling SMB3 ion Isilonwge S wf wzxyxsr

0
0
TRe: Isilon Continuous Availability bug + Disabling SMB3 ion Isilonwge S wf wzxyxsr
posted by Steve Strong on July 11, 2017, 11:10 a.m. (2 days ago)
 Ex a were 

Sent from my iPhone

On Jul 11, 2017wxwxsyk, at 5:52 AM, julian j <u@gmail.iySRW

Isilon-peeps,
W
We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementationvjd of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

To disable SMB3 entirely:
 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
 Ex a were 

Sent from my iPhone

On Jul 11, 2017wxwxsyk, at 5:52 AM, julian j <u@gmail.iySRW

Isilon-peeps,
W
We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementationvjd of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

To disable SMB3 entirely:
 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

TRe: Isilon Continuous Availability bug + Disabling SMB3 ion Isilonwge S wf wzxyxsr

0
0
TRe: Isilon Continuous Availability bug + Disabling SMB3 ion Isilonwge S wf wzxyxsr
posted by Steve Strong on July 11, 2017, 11:25 a.m. (2 days ago)
Sorry all pocket send 

Sent from my iPhone

On Jul 11, 2017, at 11:07 AM, Steve Strong <steve.strong@cinesysinc.com> wrote:

 Ex a were 

Sent from my iPhone

On Jul 11, 2017wxwxsyk, at 5:52 AM, julian j <u@gmail.iySRW

Isilon-peeps,
W
We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementationvjd of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

To disable SMB3 entirely:
 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Sorry all pocket send 

Sent from my iPhone

On Jul 11, 2017, at 11:07 AM, Steve Strong <steve.strong@cinesysinc.com> wrote:

 Ex a were 

Sent from my iPhone

On Jul 11, 2017wxwxsyk, at 5:52 AM, julian j <u@gmail.iySRW

Isilon-peeps,
W
We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementationvjd of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:
# isi smb settings global modify --support-multichannel=no
This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

To prevent Continuous Availability session setup:
# isi_for_array s /usr/likewise/bin/lwsm stop lwwit
This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

To disable SMB3 entirely:
 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10
This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  


Julian Firminger

Senior Systems Administrator
United Broadcast Facilities
Amsterdam, The Netherlands
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

New Xeon pricing

0
0
New Xeon pricing
posted by Jean-Francois Panisset on July 11, 2017, 6:15 p.m. (2 days ago)
Ars has details on the new Xeon pricing: http://www.anandtech.com/show/11544/intel-skylake-ep-vs-amd-epyc-7000-cpu-battle-of-the-decade/7 First impressions: - nothing is affordable in the "Platinum" range, cheapest part is a 16 core 2.0GHz 8153 for $3K+ - in the "Gold" range, at around $2K there's the 14 core, 2.6GHz 140W 6132 for $2111 which kind of lines up with the 14 core, 2.6GHz 135W E5-2690v4 for $2090 Which seems a bit disappointing: for the last few Xeon generations the pricing seem to come out at "you get 2 more cores for the same price point", now it's "you get the same core count, same GHZ, higher power draw and slightly more expensive". Although the 6132 will TurboBoost to 3.7GHz for single core performance vs 3.5GHz for the E5-2690v4, but that's less useful for efficiently multi-threaded rendering applications. As an aside, "Platinum" / "Gold" / "Silver" / "Bronze" sound like options at the car wash or gym memberships. I wonder who decided those were good branding ideas. JF To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Ars has details on the new Xeon pricing: http://www.anandtech.com/show/11544/intel-skylake-ep-vs-amd-epyc-7000-cpu-battle-of-the-decade/7 First impressions: - nothing is affordable in the "Platinum" range, cheapest part is a 16 core 2.0GHz 8153 for $3K+ - in the "Gold" range, at around $2K there's the 14 core, 2.6GHz 140W 6132 for $2111 which kind of lines up with the 14 core, 2.6GHz 135W E5-2690v4 for $2090 Which seems a bit disappointing: for the last few Xeon generations the pricing seem to come out at "you get 2 more cores for the same price point", now it's "you get the same core count, same GHZ, higher power draw and slightly more expensive". Although the 6132 will TurboBoost to 3.7GHz for single core performance vs 3.5GHz for the E5-2690v4, but that's less useful for efficiently multi-threaded rendering applications. As an aside, "Platinum" / "Gold" / "Silver" / "Bronze" sound like options at the car wash or gym memberships. I wonder who decided those were good branding ideas. JF To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Nexenta - Anyone still using or considering a SDS solution from them?

0
0
Nexenta - Anyone still using or considering a SDS solution from them?
posted by Greg Whynott on July 12, 2017, 4:25 p.m. (1 day ago)
Good day,

Are you or someone you know using or considering a solution from Nexenta? If so what do you think of your experience this far?

greg
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Good day,

Are you or someone you know using or considering a solution from Nexenta? If so what do you think of your experience this far?

greg

Nexenta - Anyone still using or considering a SDS solution from them?

0
0
Nexenta - Anyone still using or considering a SDS solution from them?
posted by  on July 13, 2017, 3:20 a.m.
I know one single person who's had decent luck with Nexenta.

Everyone else ...

On 12 July 2017 at 21:23, greg whynott <greg.whynott@gmail.com> wrote:
Good day,

Are you or someone you know using or considering a solution from Nexenta? If so what do you think of your experience this far?

greg

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe



--
FramestorePeter Smith Senior Systems Engineer
LondonNew YorkLos AngelesChicagoMontral

T+44 (0)20 7344 8000 M+44 (0)7816 123009
19-23 Wells Street, London W1T 3PQ
TwitterFacebookframestore.com
https://www.framestore.com/
Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
I know one single person who's had decent luck with Nexenta.

Everyone else ...

On 12 July 2017 at 21:23, greg whynott <greg.whynott@gmail.com> wrote:
Good day,

Are you or someone you know using or considering a solution from Nexenta? If so what do you think of your experience this far?

greg

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe



--
FramestorePeter Smith Senior Systems Engineer
LondonNew YorkLos AngelesChicagoMontral

T+44 (0)20 7344 8000 M+44 (0)7816 123009
19-23 Wells Street, London W1T 3PQ
TwitterFacebookframestore.com
https://www.framestore.com/

Isilon Continuous Availability bug + Disabling SMB3 on Isilon

0
0
Isilon Continuous Availability bug + Disabling SMB3 on Isilon
posted by Zorion Terrell on July 13, 2017, 1:35 p.m.

This is good to know as we are an SMB shop planning the upgrade to OneFS 8

 

Any idea if its been patched in the OneFS 8.1 variant running on the Gen6 hardware?

 

 

Zorion Terrell

IT Manager|DHX Studios

e:zorion.terrell@dhxmedia.com

t:604-684-2363| m:604-562-5148

380 West 5th Ave

Vancouver, BC Canada V5Y 1J5

 

Email Signature_DHX_Media

 

From: studiosysadmins-discuss-bounces@studiosysadmins.com [mailto:studiosysadmins-discuss-bounces@studiosysadmins.com]On Behalf Of julian firminger
Sent: July 11, 2017 2:53 AM
To: studiosysadmins-discuss@studiosysadmins.com
Subject: [SSA-Discuss] Isilon Continuous Availability bug + Disabling SMB3 on Isilon

 

Isilon-peeps,

 

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

 

Out of this came some useful tooling around handling SMB3 that I though I would share:

 

To disable SMB3 multichannel support:

# isi smb settings global modify --support-multichannel=no

This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

 

To prevent Continuous Availability session setup:

# isi_for_array s /usr/likewise/bin/lwsm stop lwwit

This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

 

To disable SMB3 entirely:

 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10

This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

 

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  

 

 

Julian Firminger

Senior Systems Administrator

United Broadcast Facilities

Amsterdam, The Netherlands

Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 

This is good to know as we are an SMB shop planning the upgrade to OneFS 8

 

Any idea if its been patched in the OneFS 8.1 variant running on the Gen6 hardware?

 

 

Zorion Terrell

IT Manager|DHX Studios

e:zorion.terrell@dhxmedia.com

t:604-684-2363| m:604-562-5148

380 West 5th Ave

Vancouver, BC Canada V5Y 1J5

 

Email Signature_DHX_Media

 

From: studiosysadmins-discuss-bounces@studiosysadmins.com [mailto:studiosysadmins-discuss-bounces@studiosysadmins.com]On Behalf Of julian firminger
Sent: July 11, 2017 2:53 AM
To: studiosysadmins-discuss@studiosysadmins.com
Subject: [SSA-Discuss] Isilon Continuous Availability bug + Disabling SMB3 on Isilon

 

Isilon-peeps,

 

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server.  - I will post more on this if/when we know it.  

 

Out of this came some useful tooling around handling SMB3 that I though I would share:

 

To disable SMB3 multichannel support:

# isi smb settings global modify --support-multichannel=no

This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS.  Persisting sessions will continue to be multichannel until dropped / re-established.  

 

To prevent Continuous Availability session setup:

# isi_for_array s /usr/likewise/bin/lwsm stop lwwit

This will disable the Isilon's file-witness service (on each node).  Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness.  This will effectively leave CA turned on (on enabled shares) but prevent clients from using it.  This will force reset any persisting SMB3-CA sessions.  

 

To disable SMB3 entirely:

 # isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10

This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from.  Persisting sessions will continue to be SMB3 until dropped / re-established.  

 

The last two are USE AT OWN PERIL.  And I have been advised that these changes should be done by an EMC engineer.  Especially the lwwit stop.  This can have drastic and immediate effects across all SMB sessions.  

 

 

Julian Firminger

Senior Systems Administrator

United Broadcast Facilities

Amsterdam, The Netherlands

Isilon Continuous Availability bug + Disabling SMB3 on Isilon

0
0
Isilon Continuous Availability bug + Disabling SMB3 on Isilon
posted by Isaac Guenard on July 13, 2017, 1:50 p.m.
Ugh. Very well timed warning. Thanks for the heads up Julian.



Isaac Guenard
Head of Pipeline

1751 Rue Richardson, 4400 Montreal
C: +1 514 746-9078


On Thu, Jul 13, 2017 at 1:34 PM, Zorion Terrell <Zorion.Terrell@dhxmedia.com> wrote:

This is good to know as we are an SMB shop planning the upgrade to OneFS 8

Any idea if its been patched in the OneFS 8.1 variant running on the Gen6 hardware?

Zorion Terrell

IT Manager | DHX Studios

e:zorion.terrell@dhxmedia.com

t: 604-684-2363| m: 604-562-5148

380 West 5th Ave

Vancouver, BC Canada V5Y 1J5

Email Signature_DHX_Media

From:studiosysadmins-discuss-bounces@studiosysadmins.com [mailto:studiosysadmins-discuss-bounces@studiosysadmins.com] On Behalf Of julian firminger
Sent: July 11, 2017 2:53 AM
To:studiosysadmins-discuss@studiosysadmins.com
Subject: [SSA-Discuss] Isilon Continuous Availability bug + Disabling SMB3 on Isilon

Isilon-peeps,

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server. - I will post more on this if/when we know it.

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:

# isi smb settings global modify --support-multichannel=no

This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS. Persisting sessions will continue to be multichannel until dropped / re-established.

To prevent Continuous Availability session setup:

# isi_for_array s /usr/likewise/bin/lwsm stop lwwit

This will disable the Isilon's file-witness service (on each node). Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness. This will effectively leave CA turned on (on enabled shares) but prevent clients from using it. This will force reset any persisting SMB3-CA sessions.

To disable SMB3 entirely:

# isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10

This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from. Persisting sessions will continue to be SMB3 until dropped / re-established.

The last two are USE AT OWN PERIL. And I have been advised that these changes should be done by an EMC engineer. Especially the lwwit stop. This can have drastic and immediate effects across all SMB sessions.

Julian Firminger

Senior Systems Administrator

United Broadcast Facilities

Amsterdam, The Netherlands


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Thread Tags:
  discuss-at-studiosysadmins 

0 Responses   0 Plus One's   0 Comments  
 
Ugh. Very well timed warning. Thanks for the heads up Julian.



Isaac Guenard
Head of Pipeline

1751 Rue Richardson, 4400 Montreal
C: +1 514 746-9078


On Thu, Jul 13, 2017 at 1:34 PM, Zorion Terrell <Zorion.Terrell@dhxmedia.com> wrote:

This is good to know as we are an SMB shop planning the upgrade to OneFS 8

Any idea if its been patched in the OneFS 8.1 variant running on the Gen6 hardware?

Zorion Terrell

IT Manager | DHX Studios

e:zorion.terrell@dhxmedia.com

t: 604-684-2363| m: 604-562-5148

380 West 5th Ave

Vancouver, BC Canada V5Y 1J5

Email Signature_DHX_Media

From:studiosysadmins-discuss-bounces@studiosysadmins.com [mailto:studiosysadmins-discuss-bounces@studiosysadmins.com] On Behalf Of julian firminger
Sent: July 11, 2017 2:53 AM
To:studiosysadmins-discuss@studiosysadmins.com
Subject: [SSA-Discuss] Isilon Continuous Availability bug + Disabling SMB3 on Isilon

Isilon-peeps,

We've been heavily effected by what appears to be a bug in Isilon OFSv8.0.0.x implementation of SMB3-CA extension where Win2012R2 and later servers will fail to release ephemeral socket connections to the Isilon, eventually leading to a network stack collapse on the server. - I will post more on this if/when we know it.

Out of this came some useful tooling around handling SMB3 that I though I would share:

To disable SMB3 multichannel support:

# isi smb settings global modify --support-multichannel=no

This will preserve the SMB3 dialect negotiation, authentication, CA and piping but disable support for cli/serv NIC RSS. Persisting sessions will continue to be multichannel until dropped / re-established.

To prevent Continuous Availability session setup:

# isi_for_array s /usr/likewise/bin/lwsm stop lwwit

This will disable the Isilon's file-witness service (on each node). Clients will still negotiate up to SMB3 dialect and if CA capable will attempt a CA session setup but this will fail when the client cannot find a file-witness. This will effectively leave CA turned on (on enabled shares) but prevent clients from using it. This will force reset any persisting SMB3-CA sessions.

To disable SMB3 entirely:

# isi_gconfig registry.Services.lwio.Parameters.Drivers.srv.MaxSmb2DialectVersion=2.10

This will stop the Isilon offering SMB3 dialect on session setup so the client side will have only 2.1, 2.0 or 1.0 to choose from. Persisting sessions will continue to be SMB3 until dropped / re-established.

The last two are USE AT OWN PERIL. And I have been advised that these changes should be done by an EMC engineer. Especially the lwwit stop. This can have drastic and immediate effects across all SMB sessions.

Julian Firminger

Senior Systems Administrator

United Broadcast Facilities

Amsterdam, The Netherlands


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Viewing all 3749 articles
Browse latest View live




Latest Images