For what it’s worth: an alleged SEACOM cable break a few kilometers from Alexandria – eta of 2.5hrs or so…
๏ปฟ
Category: General
-
#Twitter taking #WorldCup strain during matches like #FRA v #MEX
http://status.twitter.com/ is reporting the High Error Rate on Twitter.com since yesterday already, and #Trends have been temporarily disabled since the 15th of June.
On Monday the 14th of June, #Twitter blamed the site availability issues on the failed enhancements to their #Timeline Cache. At that stage, it took them a good 5 hours to restore the site to full status, but there were ongoing hassles, dragging on into the next day – at that stage, Twitter #Places needed to be switched off.
June shows 5 hours of #downtime, with 98.64% availability of the site.
Under #WorldCup traffic, matches like #France vs #Mexico, #Uruguay vs #South Africa, and other #CapeTown or #SoccerSCity matches are causing a huge amount of traffic and tweets, specifically due to the #Twitter/Worldcup link… Trying to #pepper this post with #keywords about #uptime, #availability, #highavailability and #twitter overall.
How many hashes does it take? ๐
-
Google services now hosting in South Africa and whois.co.za
Google.co.za is doing it, Gmail is doing it — resolving to 196.23.168.147 – hosted by Internet Solutions, and reverse-IP’ing to za196-23-168-147.cache.googlevideo.com (which suggests the implemenation of local youtube caching (youtube.co.za is in the offing for some time now)) – full rundown here. Youtube.co.za — registered by MarkMonitor on behalf of Google – was last updated on 15 May 2010, so it’s not far away – well, at the moment, it’s just redirecting to http://www.youtube.com/?gl=ZA&hl=en-GB They weren’t that lucky with gmail.co.za though ๐
While we’re on domain names – did you know that whois.co.za is no longer functional? You need to browse directly to http://co.za/whois.shtml – I’ll put a little something together for that…
Just kind of interesting ๐
-
SEACOM link was down again, stable now
“SEACOM’s backhaul service provider link is down from Mtunzini to Johannesburg due to an unusual failure on both primary and secondary routes due to circumstances beyond our control. SEACOM is in the process of reprovisioning customers onto a tertiary route.” is what SEACOM announced today. Some ISPs confirmed that SEACOM connectivity was restored at around 15:24 SAST, though links were only stable around 16:20 SAST.
-
Back from Tanzania and Kilimanjaro 2010!!
Made it back safe and sound from Tanzania — an amazing country!
Went all the way to the top of Mt Kilimanjaro (Uhuru Peak, 5895m) on Jan 30, 2010 at 06:10am :o)
Pictures are up at http://www.kilimanjaro2010.co.za – more narrative will follow at some stage
Yay!!!!
-
Now on Technorati too
Wow, that took long! The authority process does take its automatin’ time!
And the count-down to Kilimanjaro is on — only a few more days to go to before heading to the top – and Tanzania! ๐
-
Y2K vs 2010 vs 2016 — year rollover still an issue
Ain’t that a kick in the head! Rollover from 2009 to 2010 has not been processed correctly on many systems, including Symantec’s Endpoint Protection Manager, they have confirmed in a statement
An issue has been identified in the Symantec Endpoint Protection Manager (SEPM) server whereby all types of SEP definition content [AV/AS, IPS] with a date greater than December 31, 2009 11:59pm are considered to be โout of dateโ.
Remember that we spoke about Symantec in the glorious (dodgey) Virus Scanner Comparison.
Australian POS systems have jumped from 2009 to 2016 (pity about the malformed
dec2hexfunction they must have been using (or forgot to use?), and up to 30mil users in Germany can’t use their bank cards at ATMs or POS machines due to invalid date calculations.In addition, there also seems to be the occasional SMS (or text message) running through networks which have issues with future-stamped timestamps, causing a wide variety of reporting and integrity issues…
Hm, I wonder what implications the 2011-2012 rollover will bring with it (cue Mayan death-drone…)
-
Happy 2010!
Another year, another at leastย 730-odd opportunities! All the best for the next 8760 hours! ๐
-
news.bbc.co.uk looping includes, broken
The power of bad includes It seems that a looped include error has broken http://news.bbc.co.uk, creating repetitive nestings — or just a whole whack of emphasis on the Copenhagen climate change story which is self-included 5 times – no wonder the poor BBC Apache server reports “[an error occurred while processing this directive]” — seems that there is an incorrect citing or link… ho hum… happens to all of us ๐
I’m sure it’ll be fixed soon (though the last update time points to either a serious error or a very slow news day Blair: ‘Right to remove’ Saddam is competing directly with the Climate change prosters march in Copenhagen story (which incidentally does not have a link on the page…)
The server decries the bad code Ah, the wonder of automation… I wonder what impact this has on twitter and news.google.com feeds?
Update: 9 minutes later, they fixed it ๐
-
MTN Business goes down – Network Outage over
Seems as though systems are being restarted / links restored, 1hr 10 min later… Whole of South Africa was affected by the network outage.
Running commentary here.
To quote the WebAfrica forum “Is this “network issue” anything to do with the power failure (and generator backup failure) that occurred at MTN moments ago bringing down the entire Hetzner hosting facilities?”
mybroadband.co.za is still struggling to come back online, and has been taken offline again…
Update:ย “According to MTN Businessโ Edwin Thompson an Eskom power failure at 14:30 meant that the data centre relied on generator power this afternoon.ย Unfortunately the main generator โblew itself upโ which impacted the electricity infrastructure at the facility.ย While the power and generator system for the facility is fully redundant, Thompson explained that this it is not really possible to prepare for these types of unexpected events.
“The Gallo Manor data centre is a key hosting facility in South Africa with numerous large clients like iBurst and Hetzner.ย The downtime therefore does not only affect websites like MyBroadband, but also hosted services like authentication & billing for iBurst and hosted PBX solutions.” – mybroadband.co.za is back up…
MTN confirmed that it is experiencing network problems.ย โMTN is experiencing network problems in Pretoria and surrounding areas, which are impacting on all services. We are currently investigating the matter and we will shortly confirm how soon the problem is expected to be resolved,โ said Sameer Dave CTO of MTN SA. -
MTN Business goes down
So MTN Business’s Datacentre is down (was Verizon Business, was UUNET). JNB issue, but whole network affected. Should sites not resolve, that’s most likely the problem. No indication of source of issue (Telkom ATM link? faulty Cisco router reload?)
We await with baited breath… Meanwhile, iol.co.za and FNB, as well as iafrica.com and Standard Bank, to name a few, are down…
The MTN Business sites are all non-responsive – all DNS not returning IPs, either… “Server not Found” issue…
Sigh…
Update:
It’s a doozy:
- ns4.iafrica.com returned (SERVFAIL)
- ns0.is.co.za returned (NORECORDS)
- ns1.coza.net.za returned (SERVFAIL)
- ns0.neotel.co.za returned (NORECORDS)
- ns2.coza.net.za returned (NORECORDS)
- ns0.plig.net returned (NORECORDS)
- ns2.iafrica.com returned (SERVFAIL)
- ns1.iafrica.com returned (SERVFAIL)
Update: Indeed, can confirm that there was a power failure in Johannesburg (JNB) (co-incidentally at the same time as the comment below). Alledgedly just been restored and network is being brought back online/troubleshot… They’re also investingating why the network in Cape Town (CPT) went down on a JNB issue…
Affects the entire network in South Africa, so browsing is going to be helluva slow…
Update: 16h30 — MTN Business Customer Centre is online again, guess they’re bringing things up step by step… But sites like mybroadband.co.za are still down…
Update: Allegedly, they were going to replace and update the Durban UPS setup today (and Cape Town, too), finishing at 13h00 (15h00), which is not to have caused the issue — no notices or information yet; there was not even a notification mail triggered (via any network).
Update: 16h41 — Iafrica.com’s back, as is Standardbank.co.za.
Total down time on network: 1hr 10min – all systems and mail and DNS responding correctly from the Farm. Some server up north don’t seem so lucky – looks like they’re being physically restarted!?
Sigh. Let’s see what the “explanation” is.
-
Unproductive Video Conferences
Depending on how tight your handle is on the management of a video-conference, and what your expected outcomes are, there is a danger that the video-conference may become unproductive ๐ It’s all about the planned and structured way of planning a meeting…
The high-def version (so so worth it!!!) is 122MB as an MP4 file ๐
-
Serving video off the cloud
The website for Love never Dies features three videos, one of the Coney Island Waltz, and the other of the press release of the musical by Andrew Lloyd Webber. However, the video, run via the jquery plugin of an SWF video player is being served off http://cloudfront.loveneverdies.com. Which is in the cloud. (more…)
-
Twitter Weekly Updates for 2009-09-13
- Now also on #siliconcape – http://synch.cc is represented! ๐ #
- New ClamAV / clamscan / freshclam on old Ubuntu (Feisty or the like) http://r.sven.co.za/u/17 #
-
Twitter Weekly Updates for 2009-08-30
- ยฟumop apisdn upside down? #
-
Another old favourite is back
http://www.synch.co.za is back (after a sabbatical of some time) so everything’s all happy again on the site, as well as on http://synch.co.za — in the meantime, only the URL synch.cc had been active.
It’s been a while! ๐
-
Almost forgot about the tweet!
What would a search-engine say if I didn’t add the basic link to the source of tweets (that comes back here if all goes well): http://twitter.com/svenwelzel
But you figured that out already ๐
