Net.digest summarizes helpful
technical discussions on the comp.sys.hp.mpe Internet newsgroup and
3000-L mailing list. Advice here is offered on a best-effort, Good
Samaritan basis. Test these concepts for yourself before applying
them to your HP 3000s.
Yesterday, Claude Shannon, perhaps the only graduate student in history to change the world twice while he was still in school, died at the age of 84. Shannons masters thesis dealt with sampling theory and channel communications bandwidth and more than a little bit of that work rides with every image and sound file that you download off of the Internet. But to greatly top that work, Shannon more or less single-handedly invented information theory for his doctoral work. The profundity of that work simply cant be exaggerated. Shannon later did work in artificial intelligence, machine learning and cryptography. He was an extraordinary person who simply did extraordinary things and thought extraordinary thoughts and made them simple.
I have never believed in the indispensability of individuals, particularly scientists. The truth is there to be discovered, and like all truths, it is patient. Theres almost an irrelevancy to who actually gets to a new truth first. Climate and culture are undoubtedly more important than the names of the individuals involved. If Einstein or Newton or Darwin or Shannon had never lived, we would still know pretty much everything we know now.
Nonetheless, Shannon ranked with Einstein and Newton and Darwin, and thats no small thing.
Reader comments are no small
thing, either, and I would like to hear from you. Even negative
comments are welcome. If you think Im full of it or goofed, or
a horses behind, let me know. If something from these columns
helped you, let me know. If youve got an idea for something you
think I missed, let me know. If you spot something on 3000-L and
would like someone to elaborate on what was discussed, let me know.
Are you seeing a pattern here? You can reach me at email@example.com.
One thing I stressed when I presented MPE/iX Patch Management at this years HP e3000 Solution Symposium was to always get the latest version of Patch/iX directly from the ITRC (itrc.hp.com). Do not just assume that what you have on a PowerPatch tape or Reactive Patch tape is the latest version. Based upon a thread on 3000-L in late February, unless youve downloaded a new version in the last few weeks, you could run into a problem that would cost you many hours.
Imagine the following. You are
doing a reactive patch installation using Patch/iX to tape and get
the following message when doing the update:
It appears that this problem matches the one we have investigated in SR 8606174185. In summary, if the creation of a CSLT fails under Patch/iX (e.g., due to tape errors), rerunning the step to create another CSLT produces one where START and NL may not match. A fix for Patch/iX has been developed and is currently being tested. It will be released very soon.
Pity the poor system manager above.
In this case, he had the most recent version of Patch/iX; however, he
violated another rule I have when patching MPE/iX: If anything goes
wrong up through the creation of the CSLT or stage, start over from
scratch while you can.
Last month I finished up this column with a plea to test beta patches if at all possible, particularly those that deliver functionality. CSY is understandably cautious about releasing something into the wild unless it has been tested in a reasonable number of real-world situations. In the case of beta patches created as bug fixes, there are usually enough testers; after all, it is a fix for a problem that has likely been experienced by multiple sites. In the case of functionality betas, there is a natural reluctance to be a tester of something that is not critical to the smooth functioning of your system.
Shortly after I submitted the text for last months column, Jeff Vance posted the following: We have only had ONE download for the 6.5 ABORTPROC Patch. This is not enough exposure for us to include this patch in to the upcoming 6.5 Powerpatch 3 release. Recall that this patch began on 6.0 based on customer requests mainly from the 3000-L community. Please help us understand why there has been so little interest in trying this patch on 6.5. This knowledge will help us set expectations for the upcoming SIG enhancement implementations that CSY is ramping up for. Part of CSYs commitment to Interex and to you is based on sufficient user interest that we (CSY) will get support from you in the areas of beta testing and, perhaps, even some documentation help. [Editors note: This patch also includes the enhanced INPUT CI command that allows you to do a timed reply request to the console.]
The upcoming SIG enhancement implementations Jeff referred to are the items coming out of the new, improved SIB you probably just finished voting on. Ultimately, the success of this program will be based upon the items that make it into the mainstream releases of MPE/iX. Since the ballot items are functionality enhancements and since these are the very things that have difficulty achieving a critical mass of testers it is worth considering the responses to Jeffs plea.
Patrick Santucci made an interesting observation: It could perhaps have something to do with the fact that most of us havent yet upgraded to 6.5. This is kind of a chicken-and-egg situation. We want the functionality in 6.5 that we already have in a prior release, and so we wait; but it wont show up until a critical mass of people upgrade and test the beta patch. [Editors note: my informal research confirms Patricks observation.]
Another user complained: In my case, one of the reasons that I dont download patches is that I am not comfortable with the process. First of all, this is the only situation where I have to use FTP and so it is foreign to me. Second, once the patch is on the 3000, I always have to fumble with FROMBYTE, which is also a relearning process each time I do it.
Actually, you cannot download beta patches. They have to be ordered from the RC and arrive on tape. [I realize that CSY needs feedback on betas and having to order a beta creates a call that can then be followed up on. However, I do think this process has a discouraging effect on people dealing with beta patches.] As to the process of downloading General Released patches, the Web site (itrc.hp.com) has a very good document that takes you step-by-step through the whole process of ITRC-to-PC-to-HP 3000.
For patches that are downloadable directly to any HP e3000 using FTP (ftp.itrc.hp.com), several people complained that the size often exceeds the defaults for FTP and they waste time redoing the download. For them I have one word: Patchman. Developed by Mark Bixby, Patchman (available from www.bixby.org/mark/) is a shell script that can be used to manage all your patching needs. Among other things, its a no muss, no fuss way to download patches. Patchman takes care of everything for you, including unpacking the patches once downloaded. And no need to worry about the size of the patch. See the article I wrote about Patchman for the 3000 NewsWire for more information (in the online archives of the 3000 NewsWire, or in The Best of Hidden Value & net.digest) on the slides from my Managing Patches session (www.interex.org or www.burke-consulting.com).
For those who cannot use FTP because of proxy servers and firewalls, gnu wget, ported by Lars Appel could be the solution to avoiding the intervening PC download (his original port, as well as the port of a later version by Mark Bixby, are both available at jazz.external.hp.com). Here is some of what Lars had to say about wget: My 3000 here also does not have Internet access due to a firewall. However, Im using gnu wget for MPE/iX very often, as it can pull files from the Internet by using the same Web proxy that our browsers use.
Lars said, For example,
Note that wget can be used
for a lot of other things. People are using it, for example, to get
tracking information for HP 3000 applications from the USPS, UPS and
FedEx. Check it out.
The following question was posted
to 3000-L. In the 6.5 System Software Maintenance Manual, you are
directed to backup the system as follows:
Gilles Schipper woke a few of us up with this response:
The reason one would perform the former over the latter is to ensure that all of the files residing in the SYS account be placed at the beginning of the backup. Whether that is a good thing is, in my opinion, dubious
Perhaps, in the days of half-inch, 1600-BPI magnetic tapes, it could provide for a quicker restore of files in the SYS account than, say, files in the YAHOO account. In any case, with todays tape technology, where retrieval of any group of files from any spot on DDS or DLT tape is quick, the raison-dêtre for such file placement strategy disappears - assuming there ever really was a good reason.
Thanks, Gilles. By the way, one
very important thing was left out of both examples. If you have user
volume sets you MUST explicitly list them after the DIRECTORY keyword
or else the directories for the user volume sets will NOT be
Ive written before
about the stealth directive snmp_enabled which, when set
to false, lets you use alternatives to HP JetDirect print servers.
Turns out there is another undocumented (or at least poorly
documented) directive: default_page_size. This lets you change
default paper size. The possible values are:
John Burke is the editor of the NewsWires HiddenValue and net.digest columns and has more than 20 years experience managing HP 3000s.
Copyright The 3000 NewsWire. All rights reserved.