Keith Larman
02-01-2005, 09:51 AM
I've got a reseller account (kdlarman). One of my sites has a discussion board and a small photogallery. Well, sometime late last month a single host computer apparently visited my site 17 times (.01% of total visits) but accounted for about 50% of the site's bandwidth. In other words, the site bandwidth usage doubled late last month. This caused an interesting ripple effect due to how I set the limits set in my accounts but also due to how Westhost handles these issues.
I was out of town for about 3 days. Remember, this occured late in the month. I check my mail from a friend's computer and there are a couple messages saying "90% bandwidth usage for reseller account". Oops. And a message saying I've been exceeded. Double oops. So I figure "wow, I must have something going on. I don't have *any* high usage accounts although I have gotten within 3 or 4 gig of the limits before. But this is Jan 29 or 30 at this point... So I get home finally, order the additional gig of bandwidth, and finish doing a few things I need to do (just got back). An hour or two later I sit at the computer to log in to my reseller account. Oops, it won't take my password. What?
I've got a note saying bandwidth was added. Cool. But nothing else.
So I try again. Nope.
So I submit a bug report. A few hours later I check the ticket and... It was killed. What? I'd never exceeded bandwidth before so I didn't know if the inability to log into my reseller control panel was due to exceeded bandwidth (and I did add the additional gig the day before) or if something was somehow attacking my sites or reseller account (I hadn't yet looked into where the bandwidth was going and now I couldn't even log in so I didn't know what was going on). And all my sites were down.
Okay, having a ticket killed without explanation is a *sure* way to, um, ruffle my feathers.
So I call. The nice tech said that I had added the bandwidth but they should have sent me a note saying I needed to reactivate each account. Well, sure, but I couldn't log onto my control panel. He reactivated that and I was up again. So I went through each site starting each one up. Cool. Got to the site with the problems and checked out the webalizer log. Hmmm, not really any different. So I go over to my reseller bandwidth usage and ... Yes, I was over limit even with the additional gig added.
Okay, so now at least I understand what happened. I added the gig *before* the system turned off my reseller account deactivating all my sites. But apparently it wasn't enough of an addition (I should have gone for 2 or 3) so it just turned it all off anyway. But since the sites weren't down yet, I never got the note telling me how to turn them back on. Which wouldn't have helped anyway because I wouldn't have been able to get into my reseller panel.
So anyway, this was mid-afternoon yesterday. I went over to the add-ons and added 5 gig more bandwidth. I wait. And wait. Then I call.
Okay, like Jalal in the other thread I didn't know it wasn't a dynamic automated process. Those folk went home. No bandwidth added. I called tech support and the nice guy was apologetic, but there was nothing he could do. He could add temp bandwidth to a single site, but that wouldn't do anything for a reseller. If the overall reseller account is over limit, all the sites go down. Until the accounts people can authorize the new bandwidth. They can't add bandwidth to a reseller account. Only individual account.
So all my sites were down mid afternoon through this morning. And there was *nothing* I could do about it. Even though I sent e-mails to westhost, ordered bandwidth (twice), and talked with westhost tech support (twice) before everything went into limbo for all those hours.
Of course this is a combination of me not being able to monitor things for a while, a rogue host burning up umpteen gigs of my bandwidth, along with Microsoft's new indexing spider (POS ate up 1.5 gig of bandwidth on one of my site's last month) and the fact that resellers can't have bandwidth added even in emergency situations off hours.
So in summary fwiw for Westhost. Might I suggest that when bandwidth is added per an order to check to see if there is a "limits" approaching/exceeded issue involved. I *should* have checked myself but all of this would have been avoided if when the bandwidth was added if westhost could have checked and said "Bandwidth added but you're still over limit so you need to add more -- call us". Secondly, increasingly we are all seeing issues with worms attacking various types of sites. The phpbb2 vulnerability recently, mysql, etc. have all been targets of various malicious programs. This can cause significant spikes in bandwidth. *If* it happens suddenly and uses enough of a site's allocation things can go downhill fast. In my case I'm right now looking into locating the host that was banging my site to ban his IP. I would suggest giving the 24/7 tech support more power to allocate bandwidth to reseller accounts under certain circumstances. Thirdly, in the automated messages it would be *really* helpful if for the reseller accounts they'd include the actual values. Something like "Your bandwidth levels have been exceeded. You have 25 gig and have used 26.5 gig. You have exceeded bandwidth by 1.5 gig..." That would have gotten me to upgrade to 5 gig right away and none of my problems would have occurred.
Some problems were obviously my own making. But with more communication it would have been vastly easier to get things fixed up before anything went downhill. And being notified that the my adding of additional bandwidth was insufficient to cover the overusage would have been nice. Frankly these problems are going to increase as companies like Microsoft keep putting out horrid spider software like their new search bot and with increasing problems of automated worms banging sites left and right. And with 24/7 tech support not being able to help resellers at all once the bandwidth is exceeded (although they can help individual host plans), well, that just doesn't seem quite right...
All fwiw
I was out of town for about 3 days. Remember, this occured late in the month. I check my mail from a friend's computer and there are a couple messages saying "90% bandwidth usage for reseller account". Oops. And a message saying I've been exceeded. Double oops. So I figure "wow, I must have something going on. I don't have *any* high usage accounts although I have gotten within 3 or 4 gig of the limits before. But this is Jan 29 or 30 at this point... So I get home finally, order the additional gig of bandwidth, and finish doing a few things I need to do (just got back). An hour or two later I sit at the computer to log in to my reseller account. Oops, it won't take my password. What?
I've got a note saying bandwidth was added. Cool. But nothing else.
So I try again. Nope.
So I submit a bug report. A few hours later I check the ticket and... It was killed. What? I'd never exceeded bandwidth before so I didn't know if the inability to log into my reseller control panel was due to exceeded bandwidth (and I did add the additional gig the day before) or if something was somehow attacking my sites or reseller account (I hadn't yet looked into where the bandwidth was going and now I couldn't even log in so I didn't know what was going on). And all my sites were down.
Okay, having a ticket killed without explanation is a *sure* way to, um, ruffle my feathers.
So I call. The nice tech said that I had added the bandwidth but they should have sent me a note saying I needed to reactivate each account. Well, sure, but I couldn't log onto my control panel. He reactivated that and I was up again. So I went through each site starting each one up. Cool. Got to the site with the problems and checked out the webalizer log. Hmmm, not really any different. So I go over to my reseller bandwidth usage and ... Yes, I was over limit even with the additional gig added.
Okay, so now at least I understand what happened. I added the gig *before* the system turned off my reseller account deactivating all my sites. But apparently it wasn't enough of an addition (I should have gone for 2 or 3) so it just turned it all off anyway. But since the sites weren't down yet, I never got the note telling me how to turn them back on. Which wouldn't have helped anyway because I wouldn't have been able to get into my reseller panel.
So anyway, this was mid-afternoon yesterday. I went over to the add-ons and added 5 gig more bandwidth. I wait. And wait. Then I call.
Okay, like Jalal in the other thread I didn't know it wasn't a dynamic automated process. Those folk went home. No bandwidth added. I called tech support and the nice guy was apologetic, but there was nothing he could do. He could add temp bandwidth to a single site, but that wouldn't do anything for a reseller. If the overall reseller account is over limit, all the sites go down. Until the accounts people can authorize the new bandwidth. They can't add bandwidth to a reseller account. Only individual account.
So all my sites were down mid afternoon through this morning. And there was *nothing* I could do about it. Even though I sent e-mails to westhost, ordered bandwidth (twice), and talked with westhost tech support (twice) before everything went into limbo for all those hours.
Of course this is a combination of me not being able to monitor things for a while, a rogue host burning up umpteen gigs of my bandwidth, along with Microsoft's new indexing spider (POS ate up 1.5 gig of bandwidth on one of my site's last month) and the fact that resellers can't have bandwidth added even in emergency situations off hours.
So in summary fwiw for Westhost. Might I suggest that when bandwidth is added per an order to check to see if there is a "limits" approaching/exceeded issue involved. I *should* have checked myself but all of this would have been avoided if when the bandwidth was added if westhost could have checked and said "Bandwidth added but you're still over limit so you need to add more -- call us". Secondly, increasingly we are all seeing issues with worms attacking various types of sites. The phpbb2 vulnerability recently, mysql, etc. have all been targets of various malicious programs. This can cause significant spikes in bandwidth. *If* it happens suddenly and uses enough of a site's allocation things can go downhill fast. In my case I'm right now looking into locating the host that was banging my site to ban his IP. I would suggest giving the 24/7 tech support more power to allocate bandwidth to reseller accounts under certain circumstances. Thirdly, in the automated messages it would be *really* helpful if for the reseller accounts they'd include the actual values. Something like "Your bandwidth levels have been exceeded. You have 25 gig and have used 26.5 gig. You have exceeded bandwidth by 1.5 gig..." That would have gotten me to upgrade to 5 gig right away and none of my problems would have occurred.
Some problems were obviously my own making. But with more communication it would have been vastly easier to get things fixed up before anything went downhill. And being notified that the my adding of additional bandwidth was insufficient to cover the overusage would have been nice. Frankly these problems are going to increase as companies like Microsoft keep putting out horrid spider software like their new search bot and with increasing problems of automated worms banging sites left and right. And with 24/7 tech support not being able to help resellers at all once the bandwidth is exceeded (although they can help individual host plans), well, that just doesn't seem quite right...
All fwiw