Afrihost usage stats broken

Guys, I am getting VERY annoyed now. If this is what Afrihost support is about then it's time for me to move on. How can you mess up then not even acknowledge that you have let alone fix the problem!!!
Another email to support has yielded NOTHING>

I have work to do, I don't have the luxury of time to chase Afrihost support. Good thing I don't have a hosting package otherwise I may just be explaining to my clients why the site has been down for 5 days.
 
I think they have reached the point where an increase in subscribers necessitates bringing on more staff. That could mean managers etc as well. Normal growing pains. Hope they can keep their prices low and keep giving out 2 for 1 specials.
 
Abe said:
I think they have reached the point where an increase in subscribers necessitates bringing on more staff. That could mean managers etc as well. Normal growing pains. Hope they can keep their prices low and keep giving out 2 for 1 specials.
Click to expand...

My worry is that this is a sign of the end... I can't imagine that they can afford new staff if (as they say) they are running at a loss. I hope I'm wrong.
 
So far these are the responses I have received from AH:

Mon 1 Mar 14:10

Hi xx
I trust that you are well.:)

Could you please send through the saix & afrihost logs, as per telephonic discussion.
Have a great day & please feel free to let me know if I can be of further assistance.

Warm regards.

Faraad


Ticket Details
Ticket ID: BVY-798756
Department: Afrihost.com - One Hour Reply
Click to expand...

Tue 2 march 16:15
Hi xx.

I trust that you are well.:)

I have forwarded your request through to our server admin to look at & advise.
Have a great day & please feel free to let me know if I can be of further assistance.

Warm regards.

Faraad


Ticket Details
Ticket ID: BVY-798756
Department: Afrihost - ADSL Support
Click to expand...
Thats an interesting name for a department. I guess it's not the fixit place, just the reply place :p
 
Last edited:
Ok, I just got the first favourable reply since notifying AH of the glitch:
Hi xxxxx,

Sorry for the late response. I have alocated you another 10GB's of Bw to use while I esculate this. Hope that helps for now.

Bren
Click to expand...
At least that lets me get online with AH in the mean time while it is sorted... on second thoughts, after looking at the dreaded usage logs, I am too nervous to even contemplate using it until it is fixed. I will continue with my GConnect bandwidth for now :)
 
Last edited:
MrsBBL. before 12pm disconnect and then reconnect next day on the last day of the month. then check if you have magical overuse. i think the forum person ran away or just doesn't give a **** any more.
 
Necuno said:
MrsBBL. before 12pm disconnect and then reconnect next day on the last day of the month. then check if you have magical overuse. i think the forum person ran away or just doesn't give a **** any more.
Click to expand...

The soup thickens. Yesterday I was given 'Another 10Gb bandwidth' as stated above. The usage stats showed:
15Gb total, 10Gb available (exactly 10).

Now I haven't used this account (deliberately - I don't yet trust that it's working) and today, the usage stats show:
15Gb total, 6.018 available (8.9 used)... now what am I supposed to make of that? If I look at my current session info, there is nothing since the 1st. If I look at the raw logs, it shows 30Mb usage.

Having said all that, I do use this account occasionally for - VERY LITTLE other than email - at another premesis which accounts for the 30Mb. I'm stating this so I don't have someone at AH saying 'BUT THERE IS ANOTHER SESSION AT ANOTHER NUMBER".... without actually checking it's usage stats. Either way, the usage stats are different today from yesterday by 3.9Gb with 30Mb no usage showing in the raw logs / saix logs / session info. Perhaps they are testing? Perhaps they now took off the 8.x Gb that I was supposedly 'over' and have left me with some bandwidth? Who knows...
I better not be charged EXTRA for that usage.

Either way I am not using the account until it is sorted (with exception of the remote email machine).
 
Last edited:
New update from Afrihost

I just got email from Afrihost and I am still very confused... none of it adds up in my mind...

From: Faraad - Afrihost.com [mailto:[email protected]]
Sent: Wednesday, March 03, 2010 2:06 PM
To: xxxxx
Subject: [#BVY-798756]: ADSL CAP reached

Hi xx.

I trust that you are well.:)

As per our server admin:

On 2010-02-26 at 00:18:56 the hourly update for [email protected]
reflected a usage for the hour of 5001 Mb. This was correctly processed by
our back end system, which later calculated the usage to be over the limits,
and ordered the disconnection. This 5001Mb was incorrectly processed as
2048Gb by our front-end system for accounting purposes, which caused a
noticeable difference of around 3Gb.

The ordering of the log entries is an artefact of the way we process the
entries for billing. The duplication of entries is the result of our efforts
to resolve this query.
Have a great day & please feel free to let me know if I can be of further assistance.

Warm regards.

Faraad

Ticket Details
Ticket ID: BVY-798756
Department: Afrihost - ADSL Support
Click to expand...

So I replied...

Faraad, apologies if I am missing something – I don’t think I understand what is being said – are you saying that something was done incorrectly on Afrihost’s part and I did indeed not exceed my usage by 8Gb as stated on the usage stats page, and it has been fixed?

Also – correct me if I am wrong:
5001 Mb = 5Gb
2048 Gb = 2 Terabytes (or did you mean 2048Mb?)

I’m still not sure how I ended up 8Gb over the limit in all these figures (5Gb usage, 2Gb accounted for, 3Gb not accounted for?)
Duplication of log entries are not a worry for me, what is a worry is:

The Afrihost raw logs for the period 25 Feb to 1 Mar show in excess of 31Gb usage, the SAIX logs show 19.5Gb for the same period (12Gb difference right there)
The Afrihost session information for the period 25 Feb to 1 Mar shows usage of 18.3 Gb – corresponding to the SAIX logs.

Clearly the Afrihost raw log is incorrect as both the SAIX log and the Afrihost Session disagree with it.
Do you agree?

Apologies for continuing with this but I don’t want to use the system if it is still not fixed, let me know.

Thanks
xx
Click to expand...

Will keep bumping the thread with news
 
Last edited:
what does saix log say?

<tinfoil>

are we using a pool collective at afrihost ?

i buy 5 gig you buy 7 gig. total pool 12. i use 6 instead of just 5 and that 1 gets deducted from your 7 ?

</tinfoil>
 
Last edited:
Necuno said:
what does saix log say?

<tinfoil>

are we using a pool collective at afrihost ?

i buy 5 gig you buy 7 gig. total pool 12. i use 6 instead of just 5 and that 1 gets deducted from your 7 ?

</tinfoil>
Click to expand...

Not sure I follow with the pool collective, I imagine you aimed that at AH fundis?
In the mean time, here is another email from AH:
Hi xx.

I trust that you are well.:)

I have also asked my manager to look at, & rectify data accordingly
Have a great day & please feel free to let me know if I can be of further assistance.

Warm regards.

Faraad


Ticket Details
Ticket ID: BVY-798756
Department: Afrihost - ADSL Support
Click to expand...

Here are my stats:
SAIX
Afrihost
Afrihost Session info 25 Feb-1 Mar
 
Last edited:
benricomm [clone of ProfMartin];3666364 said:
Could this be how Afrihost can afford the low pricing per gig?
Click to expand...

No, I honestly doubt that, I do believe them when they say that they use their marketing budget to fund their services but I also think it will hurt their marketing if their adsl system isn't working well.
 
Ok, the latest update is that I received a courtesy call from Andrew or Alan (apologies if I got it wrong :)
In a nutshell what happened is:
1) Radius server went down
2) There was a glitch in the bandwidth accounting db which didn't allow the correct amount of bw to be accrued
3) The accrual was done manually and someone made an error (human error)

It does sound like there have been extenuating circumstances leading up to my experience, so all in all I understand.
It's all sorted now. I hope it was a good opportunity for Afrihost to fix a glitch or two. I honestly don't mind glitches at all and expect them in an imperfect world and I honestly want to stay with Afrihost. In future it would have been helpful to:

1) Get an acknowlegement of the glitch within an hour of logging the ticket.
2) Made a plan to get me online from the start when it became apparent that the issue lay with Afrihost.

This thread can now be put to rest from my pov and I continue to be a happy (although wiser) Afrihost subscriber.
 
MrsBBL said:
Ok, the latest update is that I received a courtesy call from Andrew or Alan (apologies if I got it wrong :)
In a nutshell what happened is:
1) Radius server went down
2) There was a glitch in the bandwidth accounting db which didn't allow the correct amount of bw to be accrued
3) The accrual was done manually and someone made an error (human error)

It does sound like there have been extenuating circumstances leading up to my experience, so all in all I understand.
It's all sorted now. I hope it was a good opportunity for Afrihost to fix a glitch or two. I honestly don't mind glitches at all and expect them in an imperfect world and I honestly want to stay with Afrihost. In future it would have been helpful to:

1) Get an acknowlegement of the glitch within an hour of logging the ticket.
2) Made a plan to get me online from the start when it became apparent that the issue lay with Afrihost.

This thread can now be put to rest from my pov and I continue to be a happy (although wiser) Afrihost subscriber.
Click to expand...

good to hear you got sorted.
 
Good to hear that AfriHost didn't try to hide the problem with excuses and ignorance, but told the client exactly what happened.
 
The_Librarian said:
Good to hear that AfriHost didn't try to hide the problem with excuses and ignorance, but told the client exactly what happened.
Click to expand...

Yes, I must admit I felt like I was getting the runaround at first - more like I got the feeling the staff weren't allowed to say what they thought lol. I could be wrong there... either way, I think I got the correct story from Alan/Andrew and honestly, that makes me happy to know they listen even if it was late.

Although I will watch stats, I believe the automated glitch at least has been sorted (I think it was related to a field type in the DB being too small for the amount of data being put into it - I was hammering the downloads and did 6Gb an hour roughly). I imagine a long type should have been a single or higher.
Before anyone raises eyebrows - that's how fast MSDN sub software comes down rofl.
 
Last edited:
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X