05-Oct-2015

Updated WordPress
Tonight I updated both blogs to WordPress 4.3.1, and consequently Aksimet to the latest version (3.1.3 comes with WordPress 4.3.1 but a new version has been released as well).
No problems whatsoever.
Next, I installed Jetpack – latest version as well – but this cannot be activated: it runs into an error:
Fatal error: Cannot find a library with slug admin-pages/class.jetpack-landing-page. in /sysblog/wp-content/plugins/jetpack/require-lib.php on line 35
The reason is that this plugin seems to require a directorypath [.INC.LIB]. That is: this does exists in the root of this plugin, and accessing it deeper into the structure, the required file does exist:
$ set def wp431:[wp-content.plugins]
$ dir

Directory WP431:[WP-CONTENT.PLUGINS]

AKISMET.DIR;1       hello.php;1         index.php;1         JETPACK.DIR;1

Total of 4 files.
$ set def [.jetpack]
$ dir .dir

Directory WP431:[WP-CONTENT.PLUGINS.JETPACK]

^.INC.DIR;1         3RD-PARTY.DIR;1     CSS.DIR;1           IMAGES.DIR;1
JSON-ENDPOINTS.DIR;1                    LANGUAGES.DIR;1     MODULES.DIR;1
SCSS.DIR;1          VIEWS.DIR;1

Total of 9 files.
$ dir [.^.INC]

Directory WP431:[WP-CONTENT.PLUGINS.JETPACK.^.INC]

FONTS.DIR;1         footer.php;1        gallery-settings.js;1
GENERICONS.DIR;1    genericons.php;1    header.php;1        IMAGES.DIR;1
IMG.DIR;1           jetpack-admin.js;1  JETPACK-ICONS.DIR;1 jetpack-jitm.js;1
jetpack-modules.js;1                    jetpack-modules^.models.js;1
jetpack-modules^.views.js;1             jetpack.js;1        jp-my-jetpack.js;1
jp.js;1             jquery^.inview.js;1 jquery^.jetpack-resize.js;1
jquery^.jetpack-sync.js;1               jquery^.spin.js;1   LIB.DIR;1
postmessage.js;1    SCSS.DIR;1          spin.js;1

Total of 25 files.
$ dir [.^.INC.lib]

Directory WP431:[WP-CONTENT.PLUGINS.JETPACK.^.INC.LIB]

ADMIN-PAGES.DIR;1   class^.color.php;1  MARKDOWN.DIR;1      tonesque.php;1

Total of 4 files.

but for some reason, this file is not reachable. It’s not a matter of protection since all directories are W:RE and files W:R so that should not be a problem..
There must be something within the plugin code that needs to be adapted before this can be used on VMS. The code where the error is generated uses a variable lib_dir , set to WP_CONTENT_DIR and additions:


if ( defined( 'ABSPATH' ) && ! defined( 'WP_CONTENT_DIR' ) ) {
define( 'WP_CONTENT_DIR', ABSPATH . 'wp-content' ); // no trailing slash, full paths only - WP_CONTENT_URL is defined further down
}

$lib_dir = WP_CONTENT_DIR . '/lib';

and somewhat further down:

$lib_dir = apply_filters( 'jetpack_require_lib_dir', $lib_dir );
$choices = array(
"$lib_dir/$slug.php",
"$lib_dir/$slug/0-load.php",
"$lib_dir/$slug/$basename.php",
);

so probably there is something wrong here that needs to be addressed. to have it working on VMS.
For the time being, I can live without this plugin but it would be nice to have it working as well.

02-Oct-2015

Router issue?
Yesterday, I had a large download – almost 6Gb – that I had to retrieve from a site that is connected through a ADSL line – which is a-symmetric – and that showed. It took about 5 hours to get the file – and that shows in the router’s statistics:

Traffic graph
Traffic graph

This morning, I had to restart the router since it had become a bit sluggish, and the NAT session table showed more sessions from my work address than are feasible; so it looks as if dropped connections are not freed where they should. Perhaps I could schedule a reboot of the router once a week, when it usually quiet…

01-Oct-2015

Maintenance
No surprises. The system is stable, the amount of memory in use is constantly about 75%, the number of processes keeps within range (typically somewhere between 60 and 65, but at times less or more, but never over a prolonged period).
Mail holds no surprises either:

Saving PMAS logs to DKA0:[LOGSARCHIVE]PMASSEP2015.zip
PMAS statistics for September
Total messages    :   2499 = 100.0 o/o
DNS Blacklisted   :      0 =    .0 o/o (Files:  0)
Relay attempts    :   1401 =  56.0 o/o (Files: 30)
Accepted by PMAS  :   1098 =  43.9 o/o (Files: 30)
 Handled by explicit rule
        Rejected :    595 =  54.1 o/o (processed),  23.8 o/o (all)
        Accepted :    194 =  17.6 o/o (processed),   7.7 o/o (all)
 Handled by content
       Discarded :    145 =  13.2 o/o (processed),   5.8 o/o (all)
    Quarantained :    138 =  12.5 o/o (processed),   5.5 o/o (all)
       Delivered :     26 =   2.3 o/o (processed),   1.0 o/o (all)

The vast majority of relay attempts was on 13-Sep, from one address and different senders (or recepients – I’ve asked Process on the layout of this logfile):

13-SEP-2015 07:41:42.43|R|59.38.97.206|rvaeh@grootersnet.nl|xiaonanzi11162@163.com|550 5.7.1 Relaying not allowed: xiaonanzi11162@1`
...
13-SEP-2015 07:42:08.01|R|59.38.97.206|ptxc@grootersnet.nl|z13699753428@vip.163.com|550 5.7.1 Relaying not allowed: z13699753428@vi`
...
13-SEP-2015 09:29:55.50|R|59.38.97.206|lpd@grootersnet.nl|z13699753428@vip.163.com|550 5.7.1 Relaying not allowed: z13699753428@vip ...

Update
On what Hunter explained, the sender at 59.38.97.206 forges the FROM: line to mimic my domain (so NONE of these messages are actually sent by me) and tries to reach the next person. Quite likely this is the offender trying to get through, or a system to be breached. So now I know the route for investigation.
End Update

Apart from these, there were a few that, according the addresses, were control-messages:

13-SEP-2015 12:08:37.00|R|202.28.68.120|admin@goodservers.com|joneslarry481@gmail.com|550 5.7.1 Relaying not allowed: joneslarry481`
13-SEP-2015 13:10:19.79|R|203.42.3.104|server@checking.net|csclus.smtp@gmail.com|550 5.7.1 Relaying not allowed: csclus.smtp@gmail.`
13-SEP-2015 14:27:37.43|R|114.43.4.43|support@microsoft.com|support@microsoft.com|550 5.7.1 Relaying not allowed: support@microsoft`
13-SEP-2015 14:33:39.42|R|196.207.30.180|cpanel@www.grootersnet.nl|arcadio.setimmi@yahoo.com|550 5.7.1 Relaying not allowed: arcadi`

That’s fine with me 🙂

Furthermore, there a a few updates to be installed, no big deal (have done it before) but it just takes time – that I need to make free.

Development system
I got a new version of nxtWare-Remote to test.
Robert’s reply on my issues was I should $ SET PROCESS/PARSE=EXTENDED before installing GNV, Java and NxtWare – and that failing to do so may cause the problems. I’m pretty sure this is set by my LOGIN.COM file – or even SYLOGIN.COM – but it is no problem to double-check.

PHP still bloating
It’s mainly during the admin pages of the blog, that WordPress spits out an error:

%HTTPD-W-NOTICED, 01-OCT-2015 13:28:02, CGI:2107, not a strict CGI response
-NOTICED-I-SERVICE, http://www.grootersnet.nl:80
-NOTICED-I-CLIENT, 37.74.59.17
-NOTICED-I-URI, POST (32 bytes) /sysblog/wp-admin/admin-ajax.php
-NOTICED-I-SCRIPT, /sysblog/wp-admin/admin-ajax.php sysblog:[wp-admin]admin-ajax.php (phpwasd:) SYSBLOG:[WP-ADMIN]admin-ajax.php
-NOTICED-I-CGI, 504850205761726E696E673A20204D6F64756C652027646F (289 bytes) PHP Warning: Module 'dom' already loaded in Unknown on line 0.
-NOTICED-I-RXTX, err:0/0 raw:1321/0 net:1088/0

Quite a number of these, and just in the same PHP source (other ones are now seldom so the reader won’t get such errors any more; Or far less).
I should probably look at this PHP source to find out what’s causing the error, perhaps it is gone after the update?? Or mention it on the WordPress site…

Only one today: It seems this script tried to output something but cannot write (of course: theer is no need to write in the directory that holds the PHP-code: So it is set ReadOnly – as it should be):

%HTTPD-W-NOTICED, 01-OCT-2015 13:36:02, CGI:2107, not a strict CGI response
-NOTICED-I-SERVICE, http://www.grootersnet.nl:80
-NOTICED-I-CLIENT, 37.74.59.17
-NOTICED-I-URI, POST (32 bytes) /sysblog/wp-admin/admin-ajax.php
-NOTICED-I-SCRIPT, /sysblog/wp-admin/admin-ajax.php sysblog:[wp-admin]admin-ajax.php (phpwasd:) SYSBLOG:[WP-ADMIN]admin-ajax.php
-NOTICED-I-CGI, 556E61626C6520746F206F70656E2027707468726561645F (71 bytes) Unable to open 'pthread_dump.log' for write, using standard output only
-NOTICED-I-RXTX, err:0/0 raw:1088/0 net:1088/0

so there is something to be changed in that script anyway…

18-Sep-2015

Moved the images of Trips, Tracks and Travels
The size of the logical disk containing the images of our “expeditions” was extended once before to 15 Gb, but even that is not sufficient; the next bit: our fall trip into Germany (2 years ago – it’s finally done) won’t fit, let alone what is still to come (there still is some work to be done). I do have spare 32Gb disks available so I decided to $ INIT one of them and move the whole directory tree containing the images over to a new directory on that disk, and add the latest album. To access them, I added this directory into the logical that contains all that is part of this environment (including the blog), leaving the now (hopefully) obsolete tree intact:

$&nmsp;sho&nmsp;log&nmsp;tracks
&nmsp;&nmsp;&nmsp;"TRACKS"&nmsp;=&nmsp;"web_disk2:[public.trtrtr.]"&nmsp;(LNM$SYSTEM_TABLE)
&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;=&nmsp;"web2016:[trackimages.]"&nmsp;&nmsp;
&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;=&nmsp;"web_disk2:[blogs.tracks.]"
&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;&nmsp;=&nmsp;"wp43:"
1&nmsp;&nmsp;"WP43"&nmsp;=&nmsp;"web_disk2:[private.wp43.]"&nmsp;(LNM$SYSTEM_TABLE)

Because of this, there was no need at all to add anything in the WASD configuration because all that is referred to is this logical configuration:
redirect /tracks /tracks/index.php
map /tracks**/ /tracks*/index.php
exec+ /tracks/**.php* (phpwasd:)/tracks/*.php* ods=5
pass /tracks/000000/* /tracks/* ods=5 search=none dir=noaccess
pass /tracks/* /tracks/* ods=5 search=none dir=noaccess

Of course, this double definition of the image directory isn’t needed at all. Since all data is copied, the old environment can simply be made inaccessible, by one simple DCL statemnent:


$ def/sys/exec tracks web2016:[trackimages.]/transs=conc, -
_$ web_disk2:[blogs.tracks.]/trans=conc, -
_$ wp43:/trans=conc

without touching anything else.

How I love the elegance of OpenVMS 🙂

Setting up a development environment
For some contrast…
I have two Personal Workstations at had: both 256 Mb, Alpha EV56 at 500 Mh – though one has been speeded up to run at the top available: 600 Mhz – it means that the PCI and memory busses are at full speed as well.
This machine got the memory of the other so it now has a whopping 512 Mb of memory 🙂
Meet Daphne: OpenVMS 8.4, booted in the cluster but with it’s own environment, compilers at all. The disk was thought to be 9Gb but turned out to be a 4.3 Gb one. It contained MySQL, SWS, WASD and WordPress versions so there was little room left. So all this was removed as well as any other data – since the system is booted into the cluster and all disks are designed to be mounted clusterwide, I can use the abundant space on Diana for storage, but have the work done on Daphne.
So far, it’s all as expected.
I could stop here and start developing, but I promised eCube half a year ago to do some testing on their development (Eclipse-based) platform; I proposed a small addition (to be abled to use it on a rooted environment where different parts of an application are stored in separate sub-trees) so I needed to install some extra software: the Nxtware Remote toolsuite. That needs Java and GNV, the toolbox to Unixify an OpenVMS environment to ease porting of *x based software more easily.
Installing the IDE (eCube Nxtware) is no problem.
Installing JAVA is no problem.
Installing GNV is no problem. Well, mostly; there are a few updates to be installed but the question is where to get them? There are several places on Sourceforge and the documentation on the site isn’t exactly helping… Installed these, but it is questionable whether these are really the lastest versions (I found newer ones yesterday…)
Configuring Nxtware shows I need to change a few system and process parameters. Process parameters for Java (otherwise it may not work), system parameters likely for the Nxtware server process. After that, configuration starts with extracting files that will be needed by either the server or clients – and there trouble starts: It cannot extract and process < .PAS> en < .COB> files; but otherwise it seems to run ton the end. Starting the server will tall….

But first, I’ll read the docs.

09-Sep-2015

Stable
This shows that after the last reboot (when the new memory was installed) the load on CPU, process slots and memory has stablelized:

Load in one week
Load in one week

So it’s only memory usage that increased up to the current level and doesn’t change very much – just a spike that is consistent with the extra processes, and so is paging: New processes require memory and of course, this must be acquired and distributed – hence tis paging.
These spikes in number of processes. memory usage and paging occur between 6:45 and 7:00, and between 18:45 and 19:00, every day. You might expect (as I did :)) that there is a batch job that runs at these moments, but $SHO QUE shows none:
$ sho que
Batch queue BACKUP, idle, on DIANA::

Printer queue DCPS_QUEUE, idle, on DIANA::"IP_RawTCP/192.168.0.251:9100",
  mounted form DCPS$DEFAULT (stock=DEFAULT)

Printer queue HPTN2100, stopped, on DIANA::"192.168.0.251:9100",
  mounted form DEFAULT

Batch queue NORMAL, idle, on DIANA::

Generic batch queue SYS$BATCH

  Entry  Jobname         Username             Status
  -----  -------         --------             ------
    218  PreciseMail Stats
                         SYSTEM               Holding until  9-SEP-2015 15:05:00
    207  PreciseMail Notify
                         SYSTEM               Holding until  9-SEP-2015 15:30:00
    190  reset_syslog    SYSTEM               Holding until 10-SEP-2015 00:00:00
    191  scan_log        SYSTEM               Holding until 10-SEP-2015 00:00:00
    192  do_backup       SYSTEM               Holding until 10-SEP-2015 00:00:00
    195  PreciseMail Nightly
                         SYSTEM               Holding until 10-SEP-2015 00:05:00
    200  PreciseMail AutoUpdate
                         SYSTEM               Holding until 10-SEP-2015 02:00:00
    810  save_logs       SYSTEM               Holding until  1-OCT-2015 01:00:00

Generic printer queue SYS$PRINT

Batch queue SYS$START, stopped, on DIANA::

Batch queue T4$BATCH, available, on DIANA::

  Entry  Jobname         Username             Status
  -----  -------         --------             ------
    143  T4$COLLECT      SYSTEM               Executing
    189  T4$COLLECT      SYSTEM               Holding until  9-SEP-2015 23:59:00

Generic server queue TCPIP$SMTP

Generic server queue TCPIP$SMTP_DIANA_00

Server queue TCPIP$SMTP_DIANA_01, stopped, on DIANA::, mounted form DEFAULT

Server queue TCPIP$SMTP_DIANA_1, idle, on DIANA::, mounted form DEFAULT

So now it is a matter of looking for logfiles created on these moments; perhaps $ ACCor $ ANA/AUDIT will show something 🙂
Track images moved
The images of all our trips, tracks and travels now take about 10 Gb – almost the full size of the logical disk that contains all of the web data. Adding another 1 Gb (or more) of the Lahn Wanderweg (walked in 2014), now (almost) ready for publication, would not fit. So I added another 32Gb disk to the system, initialized it and mover all the images to that physical disk. I could of course have created a logical disk of sufficient size (say 20Gb) which is, indeed, easier to move, and backup (just a single file); but I can do so on the current disk, once the files have been removed, the original logical disk could be used as well; just adding the other files to this drive as well – bit by bit – and re-create the disk-containers on that disk once more – in the appropriate size.
This move however gives me time and space to maneuver.
Another addition: Images and videos of Tessa – our Little Black Devil. Video won’t show fluently at the moment because the disk is heavily loaded (because backup of the track-images).