Hosting Issues

[RichieBoy67]

It is not very easy for a shared host to set the reverse dns up correctly. I would not be so hard on the host about this but I would really want to know what caused the database issues.

Did you get database error notifications? Can you post the exact errors?

[Flibnipktz]

Made me laugh, TLS. We've been with this host for two years, and I'm trying to figure out whether there is a real change in the level of service or if it's just my perception. If the former, then I think we may be able to spend a little less per month and see fewer issues.

You have access to URLJet tickets! Outstanding. You'll see that the responses are certainly prompt - if at times somewhat cryptic for a low-tech person. I'm interested in your take on this.

RB... thanks, good to know. Error as follows with related conversation:

Database error in vBulletin 4.2.1:

Invalid SQL:

SELECT post.postid, post.pagetext, IFNULL( user.username , post.username ) AS username, dateline
FROM post AS post
LEFT JOIN user AS user ON (user.userid = post.userid)
WHERE threadid = 11118
AND visible = 1

ORDER BY dateline ASC
LIMIT 0,250;

MySQL Error : Table './(edited)/user' is marked as crashed and should be repaired
Error Number : 145
Request Date : Sunday, August 23rd 2015 @ 02:51:48 PM
Error Date : Sunday, August 23rd 2015 @ 02:51:49 PM

And the ticket conversation.

Contact 08/23/2015 10:45
Site appears to be offline and database errors but no changes were made that I'm aware of. Multiple errors in the admin email, example below. Can you advise on our situation?



Staff 08/23/2015 10:54
Hi ,

Your user table was crashed..I've repaired and the site is running as expected...Please give a check
How would you rate this reply? Poor Excellent



Contact 08/23/2015 10:55
Outstanding. Thanks... how did this happen?
G


Staff 08/23/2015 10:55
Hello,

I'm doing a mysql repair on your database to repair the crashed tables.
It will be finished with in next 15 minutes. Now your site is loading.



Contact 08/23/2015 11:23
Thanks again, we seem to be back.

What happened?



Staff 08/23/2015 11:24
Hello,

It was a database table crash and I've repaired it for you.


Contact 08/23/2015 11:25
I understand that. What happened to make the database crash.

Staff 08/23/2015 11:27
Hello,

We had restarted the server due to some technical issues.

Contact 08/23/2015 11:31
Wouldn't a full, planned restart suggest closing any resident databases beforehand and advising your customers... as opposed to a repair and possible corruption issues?


Staff 08/23/2015 11:33
Hello,

It isn't an expected one.
Our maintenance works are always scheduled at off peak hours. I hope you understand it.


Contact 08/23/2015 11:39
I don't understand routine maintenance that involves an unexpected server restart. However, we seem to be back with no related issues.

Thanks for the quick response.

[RichieBoy67]

There are some maintenance updates that require a database restart but who knows the reason. I know there was a phpadmin update lately so maybe they were updating that.

If you do move to another host and if it is a shared host be sure to address your dns problems first because the ones with dns set up correctly for email may be few and far between. Also remember that a server move can sometimes be a real chore as well depending on the size of your site, etc.

[TheLastSuperman]

Well one thing is for sure, routine maintenance can require a restart but a table crashing usually stems from something else, we need to find out why it crashed and ensure it doesn't happen again. I've pm'd you back but I want this thread to stay on topic; Flibnipktz asked for advice so lets give him solid advice no matter what is found to be the culprit... along the way some advice can always be given then in the end reviewed to make a well informed decision . Some advice may not even apply to this situation but others who find this thread after having similar db errors (if they search and somehting in your post comes up etc) might try some advice given and it work for them the point being sometimes not all advice is good for you but for others it might be... take that with a grain of salt because it does not apply to every situation.

I say keep most of it in PM because to me... how I feel about this is even if I'm not paid for this to me it feels like a paid request being discussed in public which I know its not - this is just me being nice with only a little effort involved yet I'd rather keep some stuff in PM's and have valued members contribute and give good solid advice instead of seemingly derailing your thread! So We'll keep most of this to PM's and if something comes up that explains it and I can elaborate on that in a "tutorial" or "if this ever happens to you, this might be the cause" type of way in this thread then I certainly will .

Edit: I deleted some older pm's my inbox was full, fixed now.

[Dave]

The user table crashing is rather concerning, you should definitely make backups until you're sure that everything is stable again.

[Flibnipktz]

RB67 - You could be right, the downtime may have been entirely legit. But I don't know one way or another, and that's what irritates. Never mind the possibility of this happening in future. Agree about this being a chore. I'd prefer not to if it's not necessary, which is why I've appreciated the feedback in this thread.

TLB- PM's are fine. I'm not looking to trash any host... just want the service we're paying for. I'm amazed at how small the world is sometimes.

Dave - Thanks. I think we do have an automated backup as part of our package but may be looking at doing my own, given our situation.

[Dave]

Yep, it's always a good idea to have off-site backups.

[TheLastSuperman]

Don't forget with your host, hourly cron r1 backups are ran. I still run manual backups no matter what which is what Dave meant by off-site . Usually with backups they must be restored and if you do any type of work its best to run your own backups and restore if required instead of waiting on support tickets even if the host is lightning quick, the developer can be quicker if backups are handy .

Checking this now for you and until I know what is going on I prefer to keep it in pm's, once we know what the issue is then we spill the beans here so others can benefit from it .

I overuse smilies

[Flibnipktz]

Sounds good, TLS, PMs it is. Smileys and gifs will one day eliminate text, place your bets.

Fyi... Actually got a response from att.net's attcustomercare@att.com email this morning on our blocklist issue, in less than 20 minutes. Was not expecting that. They've confirmed the IP has been delisted and to check in 24 hours.

Also will have to make a habit of keeping a local copy of our site. Good common sense.

[TheLastSuperman]

Originally Posted by Flibnipktz

Sounds good, TLS, PMs it is. Smileys and gifs will one day eliminate text, place your bets.

Fyi... Actually got a response from att.net's attcustomercare@att.com email this morning on our blocklist issue, in less than 20 minutes. Was not expecting that. They've confirmed the IP has been delisted and to check in 24 hours.

Might be a while, planet of the apes evolution of the movies while lol... at least they subsitute for tone in the meantime! Good to hear regarding att.com removing the red flag .

Originally Posted by Flibnipktz

Also will have to make a habit of keeping a local copy of our site. Good common sense.

^ That is the first and absolutely most important of being a forum owner. Always run regular backups and even if automated RUN your own backup manually at least once per week in my personal and professional opinion. If you run a backup at least once a week and any type of automatic backups fails its easier to restore a site from a week ago then to start over from scratch because of a botched backup(s). I've seen automated backups and both manual backups corrupted before so every once in a while be sure to test the lat backup made to ensure that the backups your running are valid and can be restored.

MOST IMPORTANT THING EVER AS A FORUM OWNER IS TO RUN A BACKUP OF YOUR FILES AND DATABASE OFTEN.