Not yet; I expect to be upgrading around Labor Day. I just went to 7.2.1.2 from 7.2.1.1 (smallest upgrade I've done in a long time), and I'm hoping (but not really expecting) that 8 will bring some performance improvements and some serious fixes for the webui, which is a bit of a trash fire right now.
My name is Holly Ambler. I'm Isilon's Principal User Experience researcher. My role was created in October of 2015 to bring a dedicated focus to UX. I gotta admit, I laughed at the phrase "trash fire," and said, "Welll... he or she has a point."
We have been paying particular attention to the upgrade interface.
If you want to elaborate on the trash fire, or give me any other feedback about how me can improve, drop me a line at holly.ambler@isilon.com.
I've been looking for more ways to directly engage with our customers. After all, how will we know what we need to do if we don't talk to you? By the way-- the work I'm doing is dedicated to HOW we should design interfaces to enable task completion. That makes it different from the research conducted by our our product managers, who are more focused on the "what."
By the way, I have two open surveys. I'd love it if you can respond, if you haven't already.
We've encountered some issues (This very well could be edge cases!) - Initially just with some things with the CLI not being responsive given that 8.0 seems to finish the gradual "PAPI-fication" we've seen since OneFS 7.0. The unresponsiveness seems to be with the PAPI process not responding to requests. We can get around this via isi_classic but I've yet to get an understanding on why this is occurring. I am aware of a patch that is meant to address this in 8.0.0.0 and it's reportedly fixed in the .1 minor release
We also have encountered what appears to be a bug with the uNFS stack but this is still early in investigation and it appears to be an edge case. To clarify we have a fair number of clusters running 7.2.x in which uNFS was initially released and haven't had this same issue so this seems specific to 8.0.0.0.
There is a lot of good though to be had though. The biggie is with the FSA job now running consistently at scale, on our 8.x cluster this runs every 24 hours across over a billion LINs. From a technical standpoint the move to a fresher release of FreeBSD underneath should net us tangible good... Just haven't yet felt this as an end-user.
To summarize, I'd wait for at least a MR release to iron out some of these issues/bugs. This is a general good rule of thumb when running any shared storage appliance.
Chapter 7 of the release notes for 8.0.0.0 and 8.0.0.1 have current known issues published. As of this post the document was last updated May 26, 2016.
As an additional note, Isilon offers OneFS software as General Availability (which is currently what the 8.0 releases are classified as) and Target Code. To reach Target Code status, a release must satisfy specific criteria, which includes production time in field, deployments across all supported node platforms, and other quality metrics. This is taken from the Current Isilon Software Releases documentation, which can be found here:
Raj_la
1 Rookie
•
115 Posts
0
June 2nd, 2016 14:00
yes we did couple of weeks back - no issues until now ( NFS/SMB )
carlilek
2 Intern
•
205 Posts
0
June 2nd, 2016 16:00
Not yet; I expect to be upgrading around Labor Day. I just went to 7.2.1.2 from 7.2.1.1 (smallest upgrade I've done in a long time), and I'm hoping (but not really expecting) that 8 will bring some performance improvements and some serious fixes for the webui, which is a bit of a trash fire right now.
carlilek
2 Intern
•
205 Posts
0
June 14th, 2016 10:00
Hi Holly,
You may well regret this. I'll start working on an email.
--Ken
HollyAmbler
3 Posts
0
June 14th, 2016 10:00
Hi there, carlilek!
My name is Holly Ambler. I'm Isilon's Principal User Experience researcher. My role was created in October of 2015 to bring a dedicated focus to UX. I gotta admit, I laughed at the phrase "trash fire," and said, "Welll... he or she has a point."
We have been paying particular attention to the upgrade interface.
If you want to elaborate on the trash fire, or give me any other feedback about how me can improve, drop me a line at holly.ambler@isilon.com.
Thanks!
HollyAmbler
3 Posts
0
June 14th, 2016 11:00
Hah! No, I'm delighted.
I've been looking for more ways to directly engage with our customers. After all, how will we know what we need to do if we don't talk to you? By the way-- the work I'm doing is dedicated to HOW we should design interfaces to enable task completion. That makes it different from the research conducted by our our product managers, who are more focused on the "what."
By the way, I have two open surveys. I'd love it if you can respond, if you haven't already.
This one is about charts.
https://www.research.net/r/63TCWDK
This one is about the user experience as a whole.
Isilon user experience survey
Looking forward to learning from you!
-h
SCR5121
7 Posts
0
June 27th, 2016 22:00
We've encountered some issues (This very well could be edge cases!) - Initially just with some things with the CLI not being responsive given that 8.0 seems to finish the gradual "PAPI-fication" we've seen since OneFS 7.0. The unresponsiveness seems to be with the PAPI process not responding to requests. We can get around this via isi_classic but I've yet to get an understanding on why this is occurring. I am aware of a patch that is meant to address this in 8.0.0.0 and it's reportedly fixed in the .1 minor release
We also have encountered what appears to be a bug with the uNFS stack but this is still early in investigation and it appears to be an edge case. To clarify we have a fair number of clusters running 7.2.x in which uNFS was initially released and haven't had this same issue so this seems specific to 8.0.0.0.
There is a lot of good though to be had though. The biggie is with the FSA job now running consistently at scale, on our 8.x cluster this runs every 24 hours across over a billion LINs. From a technical standpoint the move to a fresher release of FreeBSD underneath should net us tangible good... Just haven't yet felt this as an end-user.
To summarize, I'd wait for at least a MR release to iron out some of these issues/bugs. This is a general good rule of thumb when running any shared storage appliance.
carlilek
2 Intern
•
205 Posts
0
June 29th, 2016 11:00
release notes shouldn't have chapters...
sjones51
252 Posts
0
June 29th, 2016 11:00
Hi Shamrok,
Chapter 7 of the release notes for 8.0.0.0 and 8.0.0.1 have current known issues published. As of this post the document was last updated May 26, 2016.
The known issues section can be found on page 53:
https://support.emc.com/docu65098
As an additional note, Isilon offers OneFS software as General Availability (which is currently what the 8.0 releases are classified as) and Target Code. To reach Target Code status, a release must satisfy specific criteria, which includes production time in field, deployments across all supported node platforms, and other quality metrics. This is taken from the Current Isilon Software Releases documentation, which can be found here:
https://support.emc.com/docu46145