Halifax IT systems currently down - HOW????
Halifax IT systems currently down - HOW????
Author
Discussion

pgilc1

Original Poster:

40,169 posts

226 months

Saturday 14th November 2009
quotequote all
Halifax online banking and call centre systems have been down all day, apparently they've had a 'power cut'.

Speaking as an IT person HOW is this possible???

Surely, given they're one of the biggest banks in the UK, there would be enough resilience that they would fail over to another datacentre within minutes at most???

I used to severely get my ass kicked by the CEO of the company i used to work for if the systems were down for minutes, let alone for the most of a day?

J-c

419 posts

204 months

The Black Flash

13,735 posts

227 months

Saturday 14th November 2009
quotequote all
Probably the usual. Boss wont spend money on something until it goes wrong, despite being told of the danger for years. Then when it does go wrong, everyone is expected to work their nuts off to put it right at short notice.

chimera40

7,259 posts

206 months

Saturday 14th November 2009
quotequote all
Mainly because they have cut back on IT drastically since the bust, all projects cancelled and half the staff paid off. The chickens are now coming home to roost. You will find a few banks in this position soon, at this moment in time all hell is breaking loose on the executive floors and every body and there Aunty Bessy are being called in.

They will now get absolutely hammered by regulatory for the systems going down and it will cost them millions, I think it works out at something like 10 million every hour the systems are down (depending on which ones of course). If trading is effected it is a monumental disaster.

Cue inquiry and large increase in IT spend closely followed by the other banks.

elster

17,517 posts

239 months

Saturday 14th November 2009
quotequote all
More than likely is that the maintenance staff haven't fuelled up the backup gennies.

This is the state I find most UPS systems as, no matter how large financial consequences they could create.

chimera40

7,259 posts

206 months

Saturday 14th November 2009
quotequote all
elster said:
More than likely is that the maintenance staff haven't fuelled up the backup gennies.

This is the state I find most UPS systems as, no matter how large financial consequences they could create.
???? UPS, out of fuel, a UPS is powered by batteries. Think you will find this is smoke and mirrors statement from the Bank, the Data Centre will have two completely independent phases for power, each Cab will have a UPS that will provide Approx 15-30 Minutes of power to the cab. The backup Generator is always fueled.

Sounds to me more like the Core switch has went down and the backup switch has failed, this is not an easy fix if this has happened. It should also have failed over to there DR site as soon as this happened or power was lost, sounds again like this has failed to happen as the fail over would be Core switch dependent. I can bet that they are trying to bring the DR site up right now as this will be faster than trying to find out what has caused the network failure.

It is standard policy to blame the grid if things go down, if you say your IT procedures failed you are left wide open.

elster

17,517 posts

239 months

Saturday 14th November 2009
quotequote all
chimera40 said:
elster said:
More than likely is that the maintenance staff haven't fuelled up the backup gennies.

This is the state I find most UPS systems as, no matter how large financial consequences they could create.
???? UPS, out of fuel, a UPS is powered by batteries. Think you will find this is smoke and mirrors statement from the Bank, the Data Centre will have two completely independent phases for power, each Cab will have a UPS that will provide Approx 15-30 Minutes of power to the cab. The backup Generator is always fueled.

Sounds to me more like the Core switch has went down and the backup switch has failed, this is not an easy fix if this has happened. It should also have failed over to there DR site as soon as this happened or power was lost, sounds again like this has failed to happen as the fail over would be Core switch dependent. I can bet that they are trying to bring the DR site up right now as this will be faster than trying to find out what has caused the network failure.

It is standard policy to blame the grid if things go down, if you say your IT procedures failed you are left wide open.
A big UPS is a big Generator.

Not a little piddly one, but the man size ones. They are diesel powered.

TheCarpetCleaner

7,294 posts

231 months

Saturday 14th November 2009
quotequote all
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...

eldar

25,274 posts

225 months

Saturday 14th November 2009
quotequote all
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....

elster

17,517 posts

239 months

Saturday 14th November 2009
quotequote all
eldar said:
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....
I agree. Did the maintenance engineers not have to regularly check it.

AngryApples

5,449 posts

294 months

Saturday 14th November 2009
quotequote all
Agree with sentiments above that this could easily be the first of many

Heres hoping others will learn from Halifax' (alleged) mistakes









kerrrrrrrrrching!

P924

1,272 posts

211 months

Saturday 14th November 2009
quotequote all
Seen it happen too.

Genny is religously tested each month. Run for a period of 30 minutes or so.

cue power outage, UPS kicks in, 5 minutes later genny kicks in. 40 minutes after runs out of Diesel. laugh

TheCarpetCleaner

7,294 posts

231 months

Saturday 14th November 2009
quotequote all
elster said:
eldar said:
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....
I agree. Did the maintenance engineers not have to regularly check it.
The company that installed the generators had full liability over testing, maintaining and fuelling the generators. They were also responsible for two years for failed switchovers, and had strict SLA's to adhere to if the system went down, which was remotely monitored by a monitoring centre handling it.

Obviously some numbnut didnt do their contracted job, therefore missing the fact it was fuelled and in full operational status, alongside a plum in the monitoring centre not doing their job either - or a comms failure between the two. did not care as was not my problem at the time as I worked for a different company who handled the servers and robotic units that failed once this state of the art (?!) genny went down! My problem was reconfiguring stuff afterwards.

The agency that I was working on at the time did like their outsourcing... This was the result it would seem.

elster

17,517 posts

239 months

Saturday 14th November 2009
quotequote all
TheCarpetCleaner said:
elster said:
eldar said:
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....
I agree. Did the maintenance engineers not have to regularly check it.
The company that installed the generators had full liability over testing, maintaining and fuelling the generators. They were also responsible for two years for failed switchovers, and had strict SLA's to adhere to if the system went down, which was remotely monitored by a monitoring centre handling it.

Obviously some numbnut didnt do their contracted job, therefore missing the fact it was fuelled and in full operational status, alongside a plum in the monitoring centre not doing their job either - or a comms failure between the two. did not care as was not my problem at the time as I worked for a different company who handled the servers and robotic units that failed once this state of the art (?!) genny went down! My problem was reconfiguring stuff afterwards.

The agency that I was working on at the time did like their outsourcing... This was the result it would seem.
AH so they had the maintenance contract as well. In that case fully understandable.

TheCarpetCleaner

7,294 posts

231 months

Saturday 14th November 2009
quotequote all
elster said:
TheCarpetCleaner said:
elster said:
eldar said:
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....
I agree. Did the maintenance engineers not have to regularly check it.
The company that installed the generators had full liability over testing, maintaining and fuelling the generators. They were also responsible for two years for failed switchovers, and had strict SLA's to adhere to if the system went down, which was remotely monitored by a monitoring centre handling it.

Obviously some numbnut didnt do their contracted job, therefore missing the fact it was fuelled and in full operational status, alongside a plum in the monitoring centre not doing their job either - or a comms failure between the two. did not care as was not my problem at the time as I worked for a different company who handled the servers and robotic units that failed once this state of the art (?!) genny went down! My problem was reconfiguring stuff afterwards.

The agency that I was working on at the time did like their outsourcing... This was the result it would seem.
AH so they had the maintenance contract as well. In that case fully understandable.
yes

I would have loved to be a fly on the wall in the meeting room where the person responsible was getting bking hehe

eldar

25,274 posts

225 months

Saturday 14th November 2009
quotequote all
elster said:
TheCarpetCleaner said:
elster said:
eldar said:
TheCarpetCleaner said:
I have worked in a datacentre that had just been constructed, and they had a powercut a month in.

The brand new generators did not kick in, because they were out of fuel. The contractors that were brought in to set up this thing (about the size of two large portacabins), failed to fuel it.

I would not have believed this, however I was the on-call engineer that night, and I had to co-ordinate a lot of the work.

Apparently the company that supplied it got sued for a very, very large amount of money...
Hope they didn't, if the data centre hadn't tested them.....
I agree. Did the maintenance engineers not have to regularly check it.
The company that installed the generators had full liability over testing, maintaining and fuelling the generators. They were also responsible for two years for failed switchovers, and had strict SLA's to adhere to if the system went down, which was remotely monitored by a monitoring centre handling it.

Obviously some numbnut didnt do their contracted job, therefore missing the fact it was fuelled and in full operational status, alongside a plum in the monitoring centre not doing their job either - or a comms failure between the two. did not care as was not my problem at the time as I worked for a different company who handled the servers and robotic units that failed once this state of the art (?!) genny went down! My problem was reconfiguring stuff afterwards.

The agency that I was working on at the time did like their outsourcing... This was the result it would seem.
AH so they had the maintenance contract as well. In that case fully understandable.
Still careless, did they check the tests were done or the gens would actually run the loading, run up on time etc? Apparently not!

Cappsy

1,725 posts

209 months

Sunday 15th November 2009
quotequote all
the MOD it im on has ups generators big ones in catterick

tuffer

9,009 posts

296 months

Sunday 15th November 2009
quotequote all
Having been involved in an IT Security project with HBOS in 2007 I am not surprised by this.........their IT team was a bloody mess.

Allanv

3,540 posts

215 months

Sunday 15th November 2009
quotequote all
tuffer said:
Having been involved in an IT Security project with HBOS in 2007 I am not surprised by this.........their IT team was a bloody mess.
Wait until you see Lloyds IT shambles... oh wait.....

eliot

11,995 posts

283 months

Sunday 15th November 2009
quotequote all
tuffer said:
Having been involved in an IT Security project with HBOS in 2007 I am not surprised by this.........their IT team was a bloody mess.
Not the sort of comment I would put on a public forum if you are really involved in IT security.