Storagesavvy's Posts

Storagesavvy's Posts

Clarification question.... For a single UCS host, with 4 HBA Initiators, are they isolated such that Initiators A and B are for boot, and C and D are for data, without crossing?   Or do you us... See more...
Clarification question.... For a single UCS host, with 4 HBA Initiators, are they isolated such that Initiators A and B are for boot, and C and D are for data, without crossing?   Or do you use A and B for boot, but A,B,C, and D are all used for data? If they are fully isolated, one option may be to use two vArrays..  One for boot, one for data..  Assign the HBAs accordingly..  Then you create a Boot vPool and a data vPool.    Create boot LUNs and assign to single hosts, create data LUNs and assign to whole cluster. That said, ViPR has boot LUN configuration for UCS if you add in the UCS Manager and create specific compute clusters
There IS definitely a difference in block storage between 10x500 and 1x5000. There are more queues and as such with the same workload you should have shorter queues, ie: lower latency. Richa... See more...
There IS definitely a difference in block storage between 10x500 and 1x5000. There are more queues and as such with the same workload you should have shorter queues, ie: lower latency. Richard J Anderson Sr Systems Engineer, Software Defined Storage, West Globals EMC Emerging Technologies Division richardj.anderson@emc.com<mailto:richardj.anderson@emc.com> 206.229.9263 VMAX SPEED, Unified SPEED, ScaleIO SPEED, Proven EMCTA
There won't be one with that name..  It will be a generic catalog API call and you will need to pass the URI for that item in the API call.  Some of the code examples show how to query the API to... See more...
There won't be one with that name..  It will be a generic catalog API call and you will need to pass the URI for that item in the API call.  Some of the code examples show how to query the API to get a list of catalog items and URI's, and then you can send the request to the URI you want.  You can also get the URI from the URL shown in your browser when you are at the order page of a particular catalog item.
Yes, you can call catalog items to get the full workflow.  You need to get the URI for the catalog item. Check out this section of the API documentation.. ViPR 2.3 REST API Also check out ... See more...
Yes, you can call catalog items to get the full workflow.  You need to get the URI for the catalog item. Check out this section of the API documentation.. ViPR 2.3 REST API Also check out this Powershell commandlet code one of my peers created..  There are modules for calling catalog items in that as well. emccode/viprshell · GitHub Richard Anderson
The EHI is essentially an SMI-S provider.   It has it's own user/password, independent of the host itself.   Are you using the EHI user/password?  Or ActiveDirectory/Windows credentials? Make ... See more...
The EHI is essentially an SMI-S provider.   It has it's own user/password, independent of the host itself.   Are you using the EHI user/password?  Or ActiveDirectory/Windows credentials? Make sure that the Windows firewall is either disabled, or allows 5988/5989 inbound.
Check the online documentation for details on configuring the compliance rules..  https://community.emc.com/docs/DOC-49148 Look under the "Configure and Use Best Practices" section, there ... See more...
Check the online documentation for details on configuring the compliance rules..  https://community.emc.com/docs/DOC-49148 Look under the "Configure and Use Best Practices" section, there are several links that start with "Compliance:..."
Rather than a specific report, take a look at the Compliance Solution Pack.  It's purpose is to monitor your environment for configurations that differ from best practices as well as business rul... See more...
Rather than a specific report, take a look at the Compliance Solution Pack.  It's purpose is to monitor your environment for configurations that differ from best practices as well as business rules.  When the Compliance engine detects a "breach" it will generate an alert and those are viewable in the Breach Report and can also be forwarded to via SNMP Trap, Syslog, Email, etc. Default rules are included when you install the Compliance SolutionPack and you can edit them.  You can enable/disable each rule independently to limit it down to what you care about. I've attached an example of some rules that are related to the specific problem you were asking about.  These are just a few of the rules available. Richard Anderson
Well, for starters it seems like we are advertising internally, and a little externally, that Google is a great way to find EMC documentation now, which does work fantastically with the current H... See more...
Well, for starters it seems like we are advertising internally, and a little externally, that Google is a great way to find EMC documentation now, which does work fantastically with the current HTML ViPR and SRM documentation.  I noticed that for the latest release of VIPR Controller (2.3) a lot of the documentation has moved to PDF only and now I can barely find anything I need when I need it.  I don't even know what document to look in half of the time because the document itself is overly broad... For example: how do I discover a Windows host and what are the requirements to make that work?   Is the answer in the "EMC ViPR Controller 2.3 Virtual Data Center Requirements and Information Guide"?  Or in the "EMC ViPR Controller 2.3 User Interface Virtual Data Center Configuration Guide"?  It turns out that the answer is mostly in Requirements Guide, but those instructions are clearly not complete because I have yet to successfully discover a Windows host with those instructions at any customer site (separate issue I know)... In contrast, if I flip to ViPR 2.2, I can find "Add Hosts and Clusters, Windows" as a hyperlink in the root catalog and bam.. It's up..  Oh and if you search Google for that..  You get the old 2.2 or 2.0 html documentation because the 2.3 pdf docs aren't being indexed by Google. (Example: http://lmgtfy.com/?q=vipr+add+host+windows&l=1).  I've heard from my own customers that they like the HTML docs because they are easy to navigate to find what they need.   Personally, as someone who has a lot of familiarity with ViPR, I find the PDF documents frustrating because it consistently takes me longer to find what I need. That said, PDF versions of the docs, or a large PDF with multiple HTML docs bundled together is useful for reading offline.  The PDFs should always be indexable/searchable and need to allow for copy/paste to get text out of the doc to be truly useful. Another thing that we should do is have a link to the documentation from within the VIPR UI and SRM UI.. ie: for whatever version of software I'm running, (eg: VIPR 2.3 or SRM 3.6, when I click documentation in the software UI, it should open a new browser window directly to the ECN Documentation index for VIPR 2.3 or SRM 3.6, respectively). Lastly, wherever the Document Index is hosted (ie: ViPR 2.3 html index is served from http://community.emc.com ) the PDF files and any other downloadable docs need to be hosted within the same realm rather than redirecting to https://developer-content.emc.com for example (ie: the VIPR Support Matrix PDF) which causes all sorts of login and SSO issues.  And customers then have to create a new account on the Developer Network (in this case) before they can then download the PDF.  Lame! Anyway, you asked..  Moving from PDF to HTML in previous versions of SRM and VIPR docs was the right direction.  Moving away from HTML is the wrong direction. Thanks!
Plan on October for ViPR 2.3 with a patch to support XIOS 4.0SP1. Plan on November for ViPR Darth to support new features on XIO 4.0.x. Richard J Anderson Sr. Systems Engineer, Software Def... See more...
Plan on October for ViPR 2.3 with a patch to support XIOS 4.0SP1. Plan on November for ViPR Darth to support new features on XIO 4.0.x. Richard J Anderson Sr. Systems Engineer, Software Defined Solutions EMC Emerging Technologies Division richardj.anderson@emc.com<mailto:richardj.anderson@emc.com> 206.229.9263
As mentioned..  Be sure you are using either SSL and 5989 or no-SSL and 5988.  Are you sure that the CMCNE version you are using is supported with ViPR 2.3?  Check the support matrix which curren... See more...
As mentioned..  Be sure you are using either SSL and 5989 or no-SSL and 5988.  Are you sure that the CMCNE version you are using is supported with ViPR 2.3?  Check the support matrix which currently shows.. CMCNE version 11.2.1, 12.0.3, 12.1.2, 12.1.5 Are you able to telnet to the SMI-S port 5988 from a command line?
Hi JohnnyHH, ViPR SRM requires two pieces in order to integrate reporting for products that there is no SolutionPack for... 1.) Create custom collection service (which is pretty much what y... See more...
Hi JohnnyHH, ViPR SRM requires two pieces in order to integrate reporting for products that there is no SolutionPack for... 1.) Create custom collection service (which is pretty much what you already did with SNMP Devices, MIBs, Masks, etc.  I'm not positive whether you did everything necessary to populate the database, as it's hard to tell from your post, so this part may or may not need more work. 2.) Create custom reports.  When the collector was not part of a SolutionPack, then there will not be an associated reports in the FrontEnd for that data.  So you will need to create custom reports for that data. A SolutionPack is a package that includes both the collector configuration files AND corresponding canned reports.  You can create your own custom reports for ANY data that is in the database, whether a SolutionPack exists/is installed, or not, but SolutionPacks provide a set of reports that EMC has created based on what customers have asked for. hopefully that helps..  We (EMC) have a custom reporting service as well that can help if you have a specific product you need to get into SRM... What is the SMI-S and SNMP device you are trying to report on? Richard Anderson Sr. SE, SDS, EMC Enterprise West
JohnnyHH, ViPR Controller has an independent support matrix and discovery mechanism from ViPR SRM.  If, as you mentioned, your array is supported by ViPR SRM but not in the ViPR Controller Sup... See more...
JohnnyHH, ViPR Controller has an independent support matrix and discovery mechanism from ViPR SRM.  If, as you mentioned, your array is supported by ViPR SRM but not in the ViPR Controller Support Matrix, then first you should check if the 3rdParty/Cinder plug in will work for your array..  If the array is not supported by Cinder, then you would need to develop a custom driver..  Writing a driver against the Cinder API would likely be the easiest way to do that. Richard J Anderson
This can happen if there were problems with DNS, Reverse DNS, firewall rules, vShield, and some other things during the vApp deployment..  Ensure that all the VM's can ping and resolve each other... See more...
This can happen if there were problems with DNS, Reverse DNS, firewall rules, vShield, and some other things during the vApp deployment..  Ensure that all the VM's can ping and resolve each other.. You can fix the issue through Centralized Management by doing the following... 1)      Navigate to centralized management 2)      Under solution packs > Independent solution pack blocks 3)      Delete the Generic-Reports 4)      Then navigate to solution packs > EMC M&R health. 5)      Delete the Reports on the frontend server 6)      Install M&R Health SP, and push only the Reports again on the frontend host
Yes, I believe this is on the roadmap for early 2015 to have several enhancements..  I also saw a note that indicates that currently ViPR will put the journal on a different physical pool from th... See more...
Yes, I believe this is on the roadmap for early 2015 to have several enhancements..  I also saw a note that indicates that currently ViPR will put the journal on a different physical pool from the clone, within the same array if there are multiple pools, but this is done by ViPR without input from the requestor. I don't know the algorithm used as of yet.
My understanding is that specifying different locations for Journal and Clone is not supported today.. 
HDFS is another access protocol/API for the ViPR Services storage engine..  (In addition to S3, Swift, Atmos, etc).  The Hadoop compute nodes can mount viprhdfs:// filesystem and access the objec... See more...
HDFS is another access protocol/API for the ViPR Services storage engine..  (In addition to S3, Swift, Atmos, etc).  The Hadoop compute nodes can mount viprhdfs:// filesystem and access the objects that are stored in ViPR Services directly. What specifically would you like to know?
Yes, you need to have a fabric with the storage array front end ports assigned before you will be able to build a vPool.
The ViPR Support matrix shows HiCommand 7.4.1, 7.6.1, and 8.0 are supported. 7.5 might work but it’s also possible that it doesn’t work correctly for ViPR. Is it possible to try 7.6.1?
Can you identify the full version of the ViPR Controller software? Also, the full version of HiCommand and the full model# of the USPV? In the vPool configuration, did you specify "Thick"?
The short answer is Yes..  But obviously there are caveats with everything.  For example, if the application is doing something with the S3 API that is not 100% implemented in the ECS/ViPR S3... See more...
The short answer is Yes..  But obviously there are caveats with everything.  For example, if the application is doing something with the S3 API that is not 100% implemented in the ECS/ViPR S3 API head at this time, there might need to be some changes made to accommodate.  But this should be rare.