New Openvms technical journal

[labadie_1]

I am happy to see a new Openvms technical journal. And disappointed: I wanted to read the article about system service interception by Ruth Goldenberg, but it goes back to
http://h71000.www7.hp.com/openvms/journal/index.html#intercept. The same goes for the Perfdat article that interested me a lot. I gave up reading at that point.

I have not understood this code from the "Alphaserver console" article

$ b = ''f$string(''f$getenv("auto-action")')')
followed by
$ sh symb b

why not just
$ wr sys$output f$getenv("auto_action")

[labadie_1]

in the article about "Alphaserver console", I am a bit surprised to see no warning about the risks of doing "Ctrl P" to see or modify a console variable. In a Cluster, (if memory serves me) if you take more than recnxinterval seconds before typing "continue", guess what happens ?

I did that to enable remote booting of satellites when it was acceptable to reboot if I was too long
>>> set mop 1
>>> set trig 1
>>> set pwsd abc...
>>> cont

[Bart Zorn_1]

Yes, and the author also shows that he likes redundant quotes. Even if the intermediate symbol would be required, the following would also work:

$ b = f$string(f$getenv("auto_action"))

f$getenv will return a string value anyway, so the use of f$string is also redundant.

Regards,

Bart

[Ian Miller.]

some of the links are not quite corrrect in places. The links for the articles you mentioned are

http://h71000.www7.hp.com/openvms/journal/index.html#perfdat

http://h71000.www7.hp.com/openvms/journal/v8/ssi.html

[Ian Miller.]

Correction

http://h71000.www7.hp.com/openvms/journal/v8/perfdat.html

[labadie_1]

Thanks Ian (you have found it with Intellitamper or something similar ?)

I have found several typos in the .html files, it seems nobody has re-read the documents before posting them online.

[Ian Miller.]

The links are correct in other places. You should report what you have found using the "Send us your comments" link. I have found that this does result in things getting fixed.

[labadie_1]

Ian

I was going to mail to "send us your comments", but the page is correct now !

Hp seems to be pretty responsive sometimes :-)

[labadie_1]

Thread closed

[Ian Miller.]

hp have responded to my report and will be fixing the links.

[Ian Miller.]

the web page has now been fixed by hp.

[Richard J Maher]

Hi,

I originally had the same problem but, just to point out, the correct links were always available from the TOC.

Also, when Ruth says "The implementation of the mechanism limits interception to system services in executive images; that
is, system services in privileged shareable images cannot be intercepted.", let me point out that this functionality is still available from Jock Gillings' article in the V7 Technical Journal. Fake_Rtl.com will still do what you want here.

I've also taken the liberty of attaching and Inner-Mode Friendlier version of the command file. Still needs work to be called from Kernel Mode (or Exec mode ASTs) but it get's alot further.

Just define FAKE_LOAD and all the RTL routines are loaded at initialize (user-mode) time.

Regards Richard Maher