Postponed
The date I would get a new connection has been postponed until 31-Jan-2013. That is: that will be the first date that the connection can be established. So this site will be up and running a few weeks longer than originally expected. I do have the anticipated address already, and both the modem and router/firewall delivered by the new ISP. That modem offers IPV6 connectivity, but I have no idea whether the firewall offers the same facilities as the current one. Eventually, if possible, I would have all IPV4 traffic pass as-is – including the IP address by the ISP – and use the current modem. If IPV6 is not a requirement, I could even think of not using that router at all. Something to discuss with the mechanic that will install the connection (free of charge).
Testing WP35
A colleague has asked me if I could help set up a WordPress installation, using latest versions of Worpress, SWS/Apache and PHP (by Mark Berryman). The site runs on an emulated Alpha (Avanti by MSI) but there were some issues, mainly functional but performance as well. Functionality should be tackled fist, so I set up an installation on Daphne, installed SWS, MOD_PHP and next Mark Berryman’s port, followed by WordPress 3.5. There were quite a number of files that needed to be renamed, since I found that, even after $ SET PROCESS/PARSESTYLE=EXTENDED, dots in a filename are replaed by underscores – and the code expects dots….No big deal, once you know what to look for.
I used the very same setup as I had done on Diana, including this blog. First, I started with the vanilla WP installation, nothing fancy, just plain out-of-the-box, and next I created the site blog that uses this code. My colleague has confirmed that the basic WP site works, next should be the test site.
If that all works, we should check whatever can be done to speed things up. Daphne is way smaller that the environment the blog will run on, and it already has shown that the way SWS works with PHP, requires quite a lot of resources, Memory, in particular, and process slots. Showing the admin page alone, requires at least 5 SWS worker processes, all requiring a full set up of the whole environment. And PHP.INI needs to be adjusted so the maximum execution time for any script is extended – well over the default 30 seconds. I have set it to 2 minutes on Daphne and even that’s not enough (though I could create the database in that time).
This offers me the opportunity to set up a WASD environment (which also exists in Daphne) to run the blogs in WP 3.5 – and later. Because I now have a working environment, be it under SWS….
12-Dec-2012
On the verge of ISP change
In a few days, my Internet subscripotion will end – and so my connection, for an unknown period. I’m depending on my new ISP for a new connection, and I have no idea what their planning will be. Given Chrismas and New Year in the next weeks, it may take a few weeks….Of course, I could have set up a co-hosting facility, for mail in particular, but my interpretation was it could take just a few days…
We’ll see…
It will give me an opportunity to do some work on the configuration – that I intended to do anyway.
07-Dec-2012
Reboot required
Not really, in fact, but it turned outr to be the easy way out for a about 50 accesses to this blog, all LEF, and waiting for one device. SDA> SHO PROC/CHAN showed all had LDA1 “Busy Dpnd” and STOP/ID did not work at all. Stopping the webserver made no difference either. Still have to figure out what “Busy Dpnd” means (Busy I can think of, but combined with “Dpnd” – whatever that may mean….”.LDA1 is the logical device for all web-based stuff, including PHP code – both generic and blog-specific.
Anyway: I rebooted Diana, it came up like a charm within a few minutes all was in working order again. Just the test-site for a new homepage needs to be set.
Thinking afterwards: I should have crashed the system, it may have given a clue on what was wrong. I forgot.
04-Dec-2012
Disconnected…
Given the problems my ISP has with keeping the same address on the Internet connection, I have decided – on 14-Nov-2012 – to switch to another ISP, as far as Internet connectivity concerns. That also meant I have to switch the HD supplier as well, but that’s not a big deal.
Analogue and Digital broadcast, over Coax, will stay where they are, as well as telephony.
We have a one-month notice, so it means that my new ISP will enable the internet connection on 14-Dec-2012.
Yesterday, about 13:40 UTC, my ISP dropped the Internet connection, HD TV and telephone. WITHOUT WARNING. Getting into contact was impossible, so I retried this morning. It turned out that there has been a severe error in procedures: this should not have done before Dec 14th, and only Internet and HDTV should have been involved…
Telephone could be restored but it would take a few days. Internet, however, would need reactivation which would – as was said – two weeks: well after the contract was officially ended.
So I called my new ISP if they could pull instalation forward, but that was impossible.
As a result, I would be left without Internet for about two weeks. The major concern would be mail that would be undeliverable, but there is little I can do about it….Just complain how things had gone wrong.
…or not?
But when I took a look at the router tonight, I found the connection was restored, and on the previous address as well. So the connection is indeed available! As is the HD-TV connectivity. Just phone doesn’t work, yet
Looking back in the logfiles, it must have been about 9:30 that the connection was restored – around the time I had finished my phone call to the service desk. It hadn’t gone that bad, apparently.
Still, I will make a complaint because this shouldn’t have happened in the first place….
Just wondering how long this connection will last, and at what time the switch will take place – and what time it will take.
However: If I would be off line for a few days, it would give me an opportunity to do some needed updates and reorganization, without inteference with normal access. Of course I can still do the updates, but there is now more stress on timing.
Well, you can’t have everything 🙂
01-Dec-2012
Monthly maintenance
Disallowance of acccess – no matter what – for some (Chinese) networks pays off. Not the number of messages to be considered spam:
PMAS statistics for November
Total messages : 9398 = 100.0 o/o
DNS Blacklisted : 1748 = 18.5 o/o (Files: 30)
Relay attempts : 10 = .1 o/o (Files: 26)
Accepted by PMAS : 7640 = 81.2 o/o (Files: 30)
Handled by explicit rule
Rejected : 7045 = 92.2 o/o (processed), 74.9 o/o (all)
Accepted : 282 = 3.6 o/o (processed), 3.0 o/o (all)
Handled by content
Discarded : 105 = 1.3 o/o (processed), 1.1 o/o (all)
Quarantained : 174 = 2.2 o/o (processed), 1.8 o/o (all)
Delivered : 34 = .4 o/o (processed), .3 o/o (all)
that number remains high, as always (especially when filtering on content), but primarily on the number of re;ay attempts. Just four of thirty contain any data: no more than 4 blocks have been assigned, which means it’s no more than a few lines. And since the size show the numbers are small, these have been deleted (after being archived, so it is still possible to have a look).
That has been all.
Apart that, since a few weeks, there are numerous attempts to run a bit-toorent like program, accessing the main server, and, at times, on another box. It may be related to that one in particular, the router makes no distinction where the connection originates. A bit-torrent is a bit-torrent, and therefore blocked.
Update WordPress
This is a real struggle. Although it seems everything is OK, at the end there is always this ‘funny’ idea of WordPress to redirect to “http://>&site<//wptest/INDEX.PHP/INDEX.PHP“, even if I added, on Mark’s suggestion, a “map /wptest/index.php/ /wptest/” line. No matter what I try, since 2.8.2 – the version that closest resembles what I use – there is this silly redirect mapping…. Bu’t I’ll keep trying. (unless someone gives me a mapping that works….)