Category Archives: Uncategorized

Revision 516 – Added parsing of OnCommand Data for Nicknames

Posted on by

In revision 516, the basic capability to ask an OnCommand/NetApp management console for the Nickname/WWPN mapping has been added. This allows the reuse of the information it has collected by various device-specific methods to add meaning and facility to VirtualWisdom.

The ability to avoid re-polling and re-querying data essentially reduces the load to the devices polled and reduces the management effort: re-using the effort already expressed to configure one tool means that no additional effort is required — so long as the information is usable.

In this case, the same parser logic is used as was added for BNA in BNA Query later expanded to both “pools” in BNA. Like the BNA work, the OnCommand query simply reformats a query and sends it through the array of parsers to vote upon:

java -jar vict.jar --nickname=osmsql://user:pass@server:port/ --nicknameout=\VirtualWisdomData\DeviceNickname\nicknames.csv

Default user/pass are not accurate, so that still needs to be resolved. Like the BNA parser, this method hits the underlying database directly, so it needs (firewalls/filters) direct access to the server, and is vulnerable to schema changes.

Revision 515 – Added Vote02-ZoneSplit2AndOther for Zone-Voting Algorithm

Posted on by

Many customers choose to give each attached device (Server, Tape, Storage) a nickname or alias, and create their zones that way. Some customers choose to nickname the individual connections (FA or HBA ports) uniquely within the individual components of a fabric. These are very simple to pull out of a copy of the zoning information — whether it be via text files from switches dumping their zoning/VSANs, or reading the zone info replicated in BNA. Both of these methods are handled by the “nickname” option to “vict”.

Others choose to store the nicknames or aliases of the various fabric members in the names of the zones themselves; for example, the zone that allows a server “Oracle44”, via Host Bus Adapter “HBA0”, to communicate with “VMAX12” via Front-end Adapter “10aA”, that zone might be called “Oracle44_HBA0_VMAX12_10aA_prod_A” — the additional “prod_A” is a commonly used method of adding text/comments to the data.

In this revision, I added “Vote02-ZoneSplit2andother.awk” for use at companies who nickname zones with additional text, such as Server_HBA and Server_Name_FA_etc_etc

Revision 514 – WBDCopyByFilterPrefixed for UDC2FilterLaconic

Posted on by

Added a WBDCopyByFilterPrefixed.xsl that (compared to WBDCopyByFilter.xsl) prefixes the filtername to copied dashboards (sans parentheses); used with UDC2FilterLaconic.xsl, results in much shorter replicated dashboard names.

The original pair of “WBDCopyByFilter” and “UDC2Filter” together with a dashboard called “ProbeSW – Summary” and a UDC called “Array_SW” would — for the value “DMX001901”, create copies of the prototype including “ProbeSW – Summary (Array_SW = DMX001901)”.

The new pair of “WBDCopyByFilterPrefixed” and “UDC2FilterLaconic” would create copies of the prototype including “DMX001901 ProbeSW – Summary”. Clearly shorter, yet scriptable and deterministic.

Revision 513 – UDC2FilterLaconic

Posted on by

Added a UDC2FilterLaconic.xsl to create filters from UDCs with shorter names; in past, the filters created are “UDCName = UDCValue”, such as “ArraySW = DMX001901” or (in the , Filters for VirtualWisdom UDC) “Application_SW = SVCS Analysis Cluster”.

This variant only produces filters named for the value: “DMX001901” or “SVCS Analysis Cluster”. I’ve called it “Laconic” because it is clearly less verbose that its model.

Revision 512 – Vict Nickname Now Parses Cisco Device-Alias Database

Posted on by

The vict (VI Client Tool) “nickname” action now understands the output of a “show device-alias database” command on cisco.

In past, customers and users generate these files either by running the command in a plink.exe or ssh session (ie: the simplest way), or by capturing a log from their ssh session and running the command manually (ie the more-work way). Unfortunately, in the awk-based parser, replacing spaces with tabs blows away the parser, so this smarter parser should be able to handle that now.

My commit note was simply “Added basic parser for output of Cisco ‘show device-database alias’ behind VIClientTool’s –nickname= option”, and that says it all to ME, but to expand for others, this is what that means:

Recall that with multi-column CSVs, Brocade “alishow”, and Brocade “zoneshow”, a user can ignore the format of the file and just send it in with a “nickname” long-option or “N” for the short-option aficionados.

Now, if you give VICT a cisco “show device-alias database” output, it’ll make sense of that too.

The logic is the same: throw the content scattered at all the parsers, and see who gets the most results.

Although this and other parsers will get the reformatted output from a BNA ext ration (and soon, NetApp), only the basic parser ever gets results form those, so this new parser shouldn’t affect anything adversely. ..but without much forethought, the user gets another format to use.

Revision 502 – Obsessive Parser Retry

Posted on by

I’m not proud of this one. Please bear with be:

In revision 493, I wrote about how the parser for “–nickname=” actually pushes content to three separate parsers, and simply chooses the one with the best results. That was all about not trying to guess the content, but leave the guessing to the parsers. Whichever one gets the most, it wins. Too easy, and extremely scalable.

Problem is, the underlying Apache lib used to fork-off the incoming stream — to avoid downloading a file multiple times to parse it — that doesn’t always seem to work.

I put a lot of time and concern into trying to figure out why, but in the end, I just added a retry-counter.

When all parsers return a “shoot, I dunno” response, we simply run it again. And again. And again. …not so obsessive because we give up after 3 times, but you’re free to make it as psychotic/obsessive as you want.

To describe this, I verbosely wrote “add retries to the parsing so that we can thrash on a file if we need to just-get-it-done”

I promise to do better design in the future, but for now, this will only re-download a file for each full retry cycle. This doesn’t matter at all for file:// URLs, but for ftp://, bnapsql://, and http://, it will show up as multiple tries.

Revision 493 – Nickname Parses ZoneShow

Posted on by

For this revision, I wrote: “enable the –nickname= function to fork inbound content to a number of parsers; the one with the most results wins. Net result: the FAE or user needs not worry what they send to the tool, it will try to figure out what the file is. Supports user-selected columns in CSV, Brocade ZoneShow, and BNA”

What does that mean, in detail?

In past, the –nickname= fed directly into a single consumer that understand the user giving a “;WWN=x” or “;Nickname=y”, and uses those columns as input to nickname data.

Now, there are three parsers all feeding from the same resource, so even remote content (ie ftp:// and http:// URLs) is only downloaded once, but forked to many parsers. Without the user worrying about format, the three parsers try to interpret the stream to see what they can dig up. The “winning” parser is the one with the most results, effectively adapting to whatever the user sends it …of the three formats currently understood directly:

  • WWN,Nickname (WWN=x and Nickname=y are still effective)
  • Brocade zoneshow output (accuracy is challenged if the user sends a logdump of a screen-scraped output; for best results, treat the output as binary, and convert it directly using plink.exe or ssh, not a screen-capture of a log dump)
  • Brocade binary zone information from BNA versions 11 or 12

To re-iterate, the following URI types are understood:

  • http://www.example.com/file.ext
  • ftp://ftp.example.com/file.ext (anonymous FTP; have not tested user/pass)
  • file://current/directory/subdir/file.ext (same as .\current\directory\subdir\file.ext in windows)
  • file:///current/directory/subdir/file.ext (three slashes, same as \current\directory\subdir\file.ext in windows)
  • bnapsql://bna.example.com/
  • no URL: –nickname=sampleZone.zone (confirmed in testcases)

Revision 492 – second skew-check

Posted on by

It seems that someone is creating Demo Databases rather than use mine (which are cleaned up by Nick).

They’re broken. Specifically, they have extra data appended at some point. HINT: if the data in your five-minute summary ends at 2011-11-16, but your last interval is 2012-04-17, you have the broken one.

PHC now detects that, using both the packages and the final summary.