MGBrose

+Clients
  • Content count

    458
  • Joined

  • Last visited


3 Followers



About MGBrose

  • Rank
    Advanced Member

Recent Profile Visitors


262,939 profile views

MGBrose's Activity

  1. MGBrose added a post in a topic: IPBoard 4.0 HTTPS/SSL Related Requests   

     
    bfarber is offering an end user toggle to run the site under ssl possible via ipboard 3.4+ ? If so how would you create a toggle like this?
     
    Seeing a end user toggle would be an amazing feature/capability! Especially with all of the security/government snooping news lately. 
  2. MGBrose added a post in a topic: Validating members need the option to change their e-mail   

    I perfer post #3's suggestion
     
    entering a password 2x is great for verification, however the more fields you have at registration the lower your conversion rates will be. 
  3. MGBrose added a post in a topic: IPBoard 4.0 HTTPS/SSL Related Requests   

     
    I'll have to try it out then. It seems like Nexus is a really popular request in regards to https. 
  4. MGBrose added a post in a topic: IPBoard 4.0 HTTPS/SSL Related Requests   

    interesting I didnt think that was possible.
     
    Originally as an example of my setup for my site: I was thinking I could do 2 sets of urls
     
    non https --> https (2 seperate urls)
     
    upload  http://ups.mazda626.net  ----> https:// d1u9jyesu396vx.cloudfront.net css  http://sd1.mazda626.net   ---->  https:// dnma4hc22tbaa.cloudfront.net images  http://pb.mazda626.net  ----> https:// d3bjtoatzo001g.cloudfront.net js  http://sd1.mazda626.net  ----> https:// dnma4hc22tbaa.cloudfront.net   but now that you mention that, as long as it works for me I can simply use   // d1u9jyesu396vx.cloudfront.net // dnma4hc22tbaa.cloudfront.net // d3bjtoatzo001g.cloudfront.net // dnma4hc22tbaa.cloudfront.net   I'll 100% have to give that a shot! The other issue I'm seeing though is that will only work for full sitewide https encryption via config file, not via "use https for logins" because use https for logins still pulls from inline stylesheets. 
  5. MGBrose added a post in a topic: IPBoard 4.0 HTTPS/SSL Related Requests   

    I fully acknowledge that both Full site ssl via https in the config file and the current  toggle for "login using https/ssl" both work for 3.4, however we are missing a number of features in regards to both using CDN's and more granular control over SSL usage in IPBoard. I suspect that a lack of support for some of these features, especially CDN's, is making it more difficult for admins to utilize https/ssl for their communities. 
     
     
    High Priority:
    SSL doesnt work properly with CDN's. Many CDN's now support SSL including amazon cloudfront. We should have alternate https/ssl cdn text fields in the backend. Which means we need an alternate ssl urls fields for the following items within general configuration: Images URL, Upload URLS, CSS URL, JS URL, 
      SSL on login shouldnt force inline stylesheet if SSL CDN urls are available (see above). 
      SSL for admin should have its own toggle via acp. Slightly Lower Priority:
    SSL isn't optional for the user(outside of login pages if thats enabled), this is nice functionality to have for users to control individually, if they want enhanced security perhaps they could click a special icon that toggles https/ssl mode (or a user profile setting)
      IPS should consider making all https/ssl toggles (like ssl for logins, and ssl for admin) controllable via a config file to keep admins from potentially locking themselves out with those settings. 
  6. MGBrose added a post in a topic: Option for inserting full size of uploaded image?   

    very cool Rohitab, I would love to get this built in as an official IPB toggle (toggle full size attachments or something of that sort). 
  7. MGBrose added a record in IP.Board   

    Flagged as Spammer Deleted Content Bypasses ModerateCP Removed Content
    When a moderator or admin flags a spammer and delete content setting is enabled for flag spammers; That users content is completely gone forever without recovery. In my case an admin flagged a user as a spammer who was actually causing problems on the forum and I had no way to review what happened. Any deleted content from the flagged as spammer setting group should be recoverable both via backend and moderatorcp. It would be ideal to have a setting to save deleted spammer content for X amount of time similar to the way the IPB is supposed to preserve deleted posts based on the preserve deleted content. This also should be considered a bug or problem that needs to be remedied in IPB 4.0
    • 0 replies
    • 0 views
  8. MGBrose added a post in a topic: Missing Embedded videos on mobile theme?   

    try deleting your postcache, this fixed it for us.
  9. MGBrose added a post in a topic: Dedicated Server Review   

     
    Ours isn't either lol, We use about 3-4gb of our 36gb currently. This is actually one of the most expensive servers they offer, they usually have cheaper ones available, but availability seems to change almost daily. Their support is fantastic though, I have yet to receive a support response that took longer than 10mins or so max. 
  10. MGBrose added a post in a topic: IPB 3.4.4 optimization problems?   

     
    A number of boards after enabling IPSEO report slow posting times, greater than 60 seconds sometimes. To fix this disable ALL PING SERVICES from ipbseo modules. 
     
    Also check your errors, SMTP connection errors can cause hang on posting.
     
    Status Updates might be a different monster though, so i'm not familiar with that issue. 
     
    Both of these symptoms are MARJOR, and should be fixed by IPS. Try disabling your ping services(delete them all), I bet that'll fix things for you. Plus when your site is as big as yours google is crawling you all day all night, so no need for ping services. 
  11. MGBrose added a post in a topic: Dedicated Server Review   

     
     
    We've been with volumedrive for about 1.5-2 years, we jumped ship to dacentec.com recently. For similar reasons, faecesty support, faecesty admin control panel, faecesty random outages(excuse my language but its 100% true). 
     
    We switched to datacentec because rent to own, for us  means paying 50.00 a month a year from now once we own our rack. On Volumedrive for our similarly performing setup with intel E3's it was costing us around 120.00
    So when my L5520 rack is paid off a year form now and if I don't have any hardware failures, I'll be paying less than some of you renting VM's :D. 
     
     
    Current Server
    (1X) Quad Core E3-1230 (4 logical cores, 8 hyperthread cores)
    16gb Ram
    2 X 1 TB harddrives
    1GBPS port
    Rent
     
    New Server:
    (2x) Quad Core L5520 (8 logical cores, 16 hyperthreaded cores)
    36gb Ram
    2 x 128gb Crucial M4 SSD Hard Drives (Mirrored in raid1)
    1 x 1TB Storage disk (for backups) 
    1GBPS port
    12 Month Rent to Own
     
     
     
    Also VolumeDrives network Speeds is garbage, heres a benchmark I ran last week comparing the two hosts while I still own both servers. VD is wayyyyy oversold hence the horrible speeds on a "1GBPS port"
     
    Dacentec results:
    -bash-3.2# sh ./bench.sh
    CPU model : Intel® Xeon® CPU L5520 @ 2.27GHz
    Number of cores : 14
    CPU frequency : 2266.818 MHz
    Total amount of ram : 8175 MB
    Total amount of swap : 4095 MB
    System uptime : 2:39,
    Download speed from CacheFly: 16.0MB/s
    Download speed from Coloat, Atlanta GA: 50.8MB/s
    Download speed from Softlayer, Dallas, TX: 18.1MB/s
    Download speed from Linode, Tokyo, JP: 6.75MB/s
    Download speed from i3d.net, NL: 13.3MB/s
    Download speed from Leaseweb, Haarlem, NL: 5.66MB/s
    Download speed from Softlayer, Singapore: 5.19MB/s
    Download speed from Softlayer, Seattle, WA: 5.38MB/s
    Download speed from Softlayer, San Jose, CA: 17.7MB/s
    Download speed from Softlayer, Washington, DC: 42.4MB/s
    I/O speed : 106 MB/s 
    (2 x 128gb softraid 1 SSD's Mdadm)


    VolumeDrive Results:
    -bash-3.2# sh ./bench.sh
    CPU model : Intel® Xeon® CPU E31230 @ 3.20GHz
    Number of cores : 8
    CPU frequency : 3192.860 MHz
    Total amount of ram : 5109 MB
    Total amount of swap : 2559 MB
    System uptime : 19:30,
    Download speed from CacheFly: 9.23MB/s
    Download speed from Coloat, Atlanta GA: 2.26MB/s
    Download speed from Softlayer, Dallas, TX: 2.74MB/s
    Download speed from Linode, Tokyo, JP: 2.16MB/s
    Download speed from i3d.net, NL: 2.35MB/s
    Download speed from Leaseweb, Haarlem, NL: 1.83MB/s
    Download speed from Softlayer, Singapore: 1.63MB/s
    Download speed from Softlayer, Seattle, WA: 2.26MB/s
    Download speed from Softlayer, San Jose, CA: 1.89MB/s
    Download speed from Softlayer, Washington, DC: 7.66MB/s
    I/O speed : 64.1 MB/s (2 x 1Tb Softraid 1 Hard disks Mdadm)
    more related info in my dacentec mini review that compares volumdrive/dacentec ( http://www.webhostingtalk.com/showthread.php?t=1260093 )
     
    To run this script on your server you can get it here: 
    wget http://freevps.us/downloads/bench.sh sh ./bench.sh (ignore iospeeds for my result I was doing some softraid rebuilding while I did this testing).
     
    Remember when your rent/lease a server you are renting the network, so always check speeds, always check traceroutes. Any server host that is primarily cogent could mean trouble especially if you don't shape your dns routing through something like cloudflare. Cogent only networks have a ton of hops and sometimes poor speed for, dacentec is a 60-40 split, and has a really solid network for what you pay. 
     
     

     
    Dispute via paypal, youll get your money back. Even if you lose the money volumedrive isn't worth the hassle of having as a host. We get lots of random downtime with them, which is why we switched. 
     
    As for the other hosts:
    I have no direct experience with datashack or linode, but people seem to say they are decent.  Also rackspace is great, but current "cloud pricing" is ridiculous when compared to budget dedi's. Also hostgator is also good for VM's, but ridiculously  overpriced (partially because they are managed servers), I do have a client running on a small vm though on hostgator and they are very reliable/have great support. 
     
    Hope this helps!
  12. MGBrose added a post in a topic: Great Results with Nginx Cache   

    Awesome thanks Nuno! I'll have to play with this this weekend if I have the chance.
  13. MGBrose added a post in a topic: Great Results with Nginx Cache   

     
    Parisians Config?:
    So will parisians setup not yield any benefits in its current config(without guest sessions disabled)?
     
    Nunos Config?
    And would Nunos setup be considered unsecure?
     
    Nuno I notice you mention SSI in the first post and then in a later post you mentioned you never ended up setting it up? So does the SSI script from the first post work properly? Are their any drawbacks to this method of topic caching? 
     
     
     
    Getting IPS to integrate improvements:
    IPS seems to feel differently about what is and isn't a bug, if its performance improvement related, they pretty much qualify it as feedback. Post either a new IP.Board feedback thread , However it took me almost a year to get one major suggestion pushed through and confimed, and it actually got pushed through after I gave up on it because of all the community resistance(certain changes are tougher for people to want to accept) " Attachments in Fast Reply "
     
    Alternatively you can suggest that in an IPB 4.0 Performance thread that I have running that seems to have gained a little momentum:  http://community.invisionpower.com/topic/384008-ipb-40-performance-feature-requests/
     
    IPB 4.0 is our opportunity to get IPS to consider alot of performance/scaling problems were currently facing so if we can identify the best methods to implement something and it is dubbed safe/secure, I'm sure they will integrate it. They've already committed to integrating better cdn support for minified files.  I would honestly love to see nginx and varnish natively supported as a caching mechanism, with proper storage/flushing of cache controlled by ipboard core. 
     
    Disable Guest Cookie, thats it?
    Is all thats needed for a clean nginx cache is the ability to disable guest cookie? Also what problems could guests have if this cookie is removed/disabled? 
     
    Logged in User cacheable in any way(via nginx cache)?:
    Last question, what if anything can we do to nginx cache logged in users? Is their anything we can cache for them safely to improve performance for them? (the only reason I ask is because I have alot of international members, so I'd love to improve their performance as best I can as well.)
     
    Hook/Mod Able?
    Also can this be done with a simple mod? And can SSI support be done with a simple mod as well? 
     
     
    I know this is about 10+ questions, but if I can get the correct answers I can see if I can get our plugin dev[Esther with HeadStandConsulting] to quote me on how much a mod/hook to do this properly would cost.
  14. MGBrose added a post in a topic: Massive DDoS attack,please help!   

    I highly doubt your getting DDOSed after being 4 weeks old unless your running a site with some kind of illegal content. Those sites tend to get DDOSed more for moe reasons. Especially by competition.
     
    Also 4 threatening IP addresses from cloudflare is nothing, I have 2.2k "unique threats" per month, amd 83k "threat visits) per month. If 4 ips are ddossing you  simply block those IPs from connecting to your server. But a 4 IP DDOS really isn't a ddos at all. This sounds like imporper server configuration and optimization rather than an attack.
     
    We have been running our forum for 10 years and have never had a layer 7 attack that I know of. Also this should be very easy to detect with a tool like munin, where you should see an insane amount of reqests from seemingly legitimate users. Also a small layer 7 attack can easily be mitigated with varnish cache or nginx cache for guests. You really need to consult an server management expert, if you need some legitimate reccomendations pm me, and I'll send you the contact info for a few I have worked with. 
  15. MGBrose added a post in a topic: IP.Boards SMTP or PHP Mail()?   

     
    Hmm mandrill is .10 more per thousand than amazon SES. I might have to consider switching eventually if my costs get high enough.
     
    However right now, I route everything to mandrill, I hate running my own email servers so many places for problems. And Mandrill gives me shiny activity reports! 
     
    BTW we had horrible results with PHPmail on our site, SMTP proced to be way more reliable. PHPmail wasn't getting emails to alot of our members inboxes.