Web Hosting Talk : Web Hosting Main Forums : Web Hosting : Big Slow Downs....

[gowireless]

Greetings All,

I have a friend who operates a fairly successful website and they are experiencing some issues. During busy times of the day when their site is getting huge traffic, it slows WAY down (almost to the point where pages won't load).

It is my understanding that "burstable" bandwidth would benefit him. Is that true? Also, what could help him to resolve this issue?

Thanks!

[outsource]

Do you know if he/she is in a shared or dedicated hosting environment?

[Joseph]

The slowdown may be unrelated to bandwidth, but the server load. The extra load on the server may be a result of the new worms flooding mail servers, or the server could just be heavily loaded in general.

[wheimeng]

Hardly can be insufficient bandwidth issue as it happens all the time during peak hour. It could be the host has limited bandwidth limit for each site in Apache or the server load is high and resulted slow load time.

[gowireless]

To answer you Outsource, they are in a shared situation.

I don't believe that its an email issues like you mentioned Joseph, because this has been an issue for some time and not just recently. The server slows way down during specific peak times, that is why I originally thought it was a bandwidth issue.

[net-trend]

Is the server using the cPanel control panel?

[dsotmoon]

Originally posted by gowireless
The server slows way down during specific peak times,

also could be a RAM issue, what are the server specs?

[gowireless]

I'm not sure off hand if the server is using Cpanal or the server specs (sorry). Really was just trying to decifer what could possibly be slowing things down and how to help speed them up.

Can anyone tell me what makes Burstable bandwidth different from other kinds. Thanks.

[jasonl813]

Most likely reasons for slowdowns:

1.) Server on an unmetered capped port (as low as 2-3mbps with some servers)

2.) CPU is overloaded (slow CPU that can't handle many tasks like a Celeron or Duron)

3.) RAM is all used up so the server is being forced to use a swap file.

4.) The data center is seriously overselling their bandwidth. You may be able to see if this is the case by doing a traceroute to see if pings jump at any spot down the line.

Plus some others that I know am forgetting, but the above probably cover about 95% of the most likely possible causes.

[Mike_john]

Hello jasonl813

Would you please explain your these two reasons.

1.) Server on an unmetered capped port (as low as 2-3mbps with some servers)

4.) The data center is seriously overselling their bandwidth. You may be able to see if this is the case by doing a traceroute to see if pings jump at any spot down the line.

[Brightadmin]

hi

i hope jasonl813 given reasonable reasons. To solve your problem, i suggest you to talk with your service provider asap.
obtain the reasons from them. and then take necessary steps to avoid the same.

Regards

[tazd9t9]

If the site is that busy you may want to consider telling your friend to move to a dedicated server. Very busy sites on a shared server is not a great combination.

[sprintserve]

I think what everyone missed is this:

1. What's the nature of the site? Does it use scripts, DB etc? Which scripts?

2. When you say busy, what's the nature of the site statistics? Example what's the number of visitors at peak, total hits a day and more.


It may be totally unrelated to the proficiency of the host but rather a problem that's totally caused by your friend's site. Regardless of how much bandwidth or how high-end the servers are, there will also be usage that can overwhelm it. It all depends on the usage on it. So perhaps your friend's site may be the cause of the issue.

If you can provide the statistics, we can probably comment better. But generally, these plus Jason's comments will probably round it up quite nicely.

To help them speed things up, you need to get to the cause of the issue, before deciding.

[jasonl813]

Originally posted by Mike_john
Hello jasonl813

Would you please explain your these two reasons.

1.) Server on an unmetered capped port (as low as 2-3mbps with some servers)

4.) The data center is seriously overselling their bandwidth. You may be able to see if this is the case by doing a traceroute to see if pings jump at any spot down the line.

1.) Some servers are allowed a specific maximum of bandwidth such as FDC servers. If the server is busy with alot of transfers then everyone's connection is throttled down to compensate with the available bandwidth. If your site isn't the one with large file transfers, some other site on the server may have file downloads that are slowing down the connections at peak times.

4.) If you do a 'tracert yourdomain.name' from a command line, it will show the data patch from that system to the server. It shows the roundtrip response time through each router the data must pass through to get to your server. If the data center is getting too much traffic then you would see a big response time increase at the end of the tracert. You may also see some '*' which could indicate that data packets are being dropped which would also slow down your site.

[xp101]

Originally posted by jasonl813
4.) If you do a 'tracert yourdomain.name' from a command line, it will show the data patch from that system to the server. It shows the roundtrip response time through each router the data must pass through to get to your server. If the data center is getting too much traffic then you would see a big response time increase at the end of the tracert. You may also see some '*' which could indicate that data packets are being dropped which would also slow down your site.

How to read the tracert output? If I tracert my site and it returns the following, is this bad or good? I'm experiencing slowness too.

Tracing route to mysite.com [xx.xx.xxx.192]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 * * * Request timed out.
3 33 ms 31 ms 36 ms 12.244.21.97
4 61 ms 56 ms 61 ms 12.244.64.73
5 62 ms 86 ms 64 ms 12.119.199.21
6 19 ms 18 ms 14 ms gbr2-p20.st6wa.ip.att.net [12.123.44.150]
7 15 ms 20 ms 12 ms ggr1-p3100.st6wa.ip.att.net [12.123.44.133]
8 19 ms 14 ms 12 ms att-gw.sea.above.net [192.205.32.234]
9 44 ms 32 ms 29 ms so-3-2-0.mpr3.sjc2.us.above.net [208.185.175.182
]
10 43 ms 35 ms 50 ms so-2-0-0.er10a.sjc2.us.above.net [64.125.30.90]

11 44 ms 34 ms 34 ms above.sjc.assertivenetworks.net [209.66.79.101]

12 38 ms 37 ms 30 ms assertive.managed.com [66.154.102.5]
13 49 ms 44 ms 35 ms xx.xx.xxx.192

Trace complete.