Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by Dorsey

  1. I recently asked to have subversion installed on our VPS so I could use it with NetBeans. The support staff was very responsive and now my problem is "connecting" NetBeans to subversion. If anyone has any experience with this, the error I'm getting is: org.tigris.subversion.javahl.ClientException: Repository has been moved Repository moved permanently to 'http://regnavigator.com/subversion'; please relocate It might as well say: "Didn't work - try something else" for all help it provides. I've used both svnadmin and svn mkdir to create the repository at www.<domain>.com, and the directory does appear in the document root, but I just can't seem to get past this. I've also used a variety of alternatives such as plink/ssh to access the repository with similar, but not identical, results. Does anyone know what the problem is from what I've provided above? I doubt that the SVN installation is at fault.
  2. It's clear that we (you and I) have different views of your performance. Rather than challenge your points, let me say that I find your comments both offensive and condescending, undeserving of a reply. I am the customer after all, and an attitude of "do it yourself" tells me all I need to know about your organization. How embarrassing for you - I hope your response was not posted for others to see, especially potential customers. I took the time to reach out to you with honest observations about our experience, asked for a closer working relationship with TCH, even offering to pay extra for extra work, and gave you a chance to keep us as a customer - what more could a vendor want? What I got for my trouble was a very defensive reply that answered nothing. I was hoping that you'd take this as an indication of a problem that needs attention, not a nuisance that needs elimination. Your angry, vitriolic, diatribe was not exactly the reply of a businessman truly interested in customer service, wouldn't you agree?
  3. After having stuck my neck out by recommending TCH over other sites when moving from another well-known hosting service a year ago, I've learned a lot: compared to other fully managed dedicated hosting services, TCH is the near the bottom in cost, and somewhere near the middle in service. This is not a criticism, just an honest appraisal. As the saying goes: "you get what you pay for." What we like: * Cost. * A very simple process for getting up and running quickly. * Easy to use GUIs for routine administrative tasks. * Pre-emptive actions that we're probably not even aware of that keep things running smoothly. * Patience with us during the learning period. This is mutual. * Forums and Knowledge Base * Very few outages. What we don't like: * Slow level 2 response. We don't "cry wolf" or flag managers just to get attention and seldom categorize problems as urgent, but 24 hours on high priority problems is too long. * Significant changes without notice and without subsequent acknowledgement. I've been around a long time (>25years in the field), and I KNOW when something significant was changed. Just acknowledge it when it happens, I'll respect you more. * No explanation of things that have gone wrong (FTP not working) or broken (DNS down) so we can take possible pre-emptive action ourselves or modify our procedures in the future. * Seemingly few people capable of handling very serious problems. I can count the top-level support staff on the fingers of one hand. What can be done to improve service: * E-mail notifications BEFORE making significant changes, or at least give us the option of being warned if we want. This includes software and hardware upgrades, reboots, planned downtime, etc. For example, recompiling PHP invariably breaks our site - why not warn us in advance, you have our contact information? * Understand that some of us have exhausted all possible causes and solutions before turning to you. Don't force us to spend days creating test cases to prove that we're right and you're wrong. * Don't change root password without letting us know, especially if we've just changed it ourselves. When we need root password, we need it NOW, not the next day after your billing department has approved release. Again, you have the contact information. * Offer a higher-level service package (yes - at higher cost) where we have direct contact with a support team. Working some problems through junior-level screeners takes way too long. If you promise to make top-level help available, we'll promise not to waste their time.
  4. Either no one knows the answer or no one else is interested, but I've found the solution. Invoking the setPort( string ) method will access any port defined by the WSDL.
  5. I've been using MX for a year now, a I'm very happy with it. After climbing up and over the learning mountain, I can now very quickly maintain our site (PHP/MySQL model). By the way, my advice is to just work through all their tutorials, however long it takes, before a) banging your head against the wall, or buying a book on it. As for FW, I'm also reasonably happy with that, but we only use it to create nav bars and buttons. I really appreciate the ease of creating four-state roll-overs. Again, walk through the tutorials to save much time and grief. Just recently, we were able to quickly and painlessly upgrade our graphics to match other site changes without having to change a line of code. Well, many, many lines of code were changed, but MX and FW did all of it for us without a hiccup or error. Dorsey
  6. I'm accessing a sample weather web service at http://live.capescience.com/wsdl/GlobalWeather.wsdl via SOAP_WSDL. There are two bindings within that service (if this is the correct terminology): "StationInfo" and "GlobalWeather". After creating creating the proxy object, I only have access to the StationInfo methods. The StationInfo binding is defined first in the WSDL, so I don't think this is a coincidence. In other words, the following code: >require_once 'SOAP/Client.php'; $wsdl = new SOAP_WSDL('http://live.capescience.com/wsdl/GlobalWeather.wsdl'); $weather = $wsdl->getProxy(); yields an object ($weather) with only the StationInfo methods. I tried: >$weather = $wsdl->getProxy('GlobalWeather'); but that fails deep within the bowels of SOAP_WSDL. Does anyone know how to access the "other" binding? I must be missing something. Thanks for any help. A reference to an online resource with plenty of examples that explain the above would be greatly appreciated. Dorsey
  7. OK, I must have had a problem of some sort of error between the actual executing code and what I provided above, as it's now creating the SOAP_WSDL object correctly. Thanks for confirming that the sample I provided should work. I needed an independent opinion. Dorsey
  8. PHPInfo tells me that the PEAR module is installed, and the TCH help desk me that they've installed the SOAP package. My question now is: how do I use it? I'm looking for a simple example that creates a SOAP client object. Here's sample code I was provided: ============================== // Load the PEAR::SOAP client class require_once 'SOAP/Client.php'; // Read the GlobalWeather WSDL document $wsdl = new PEAR::SOAP_WSDL('http://live.capescience.com/wsdl/GlobalWeather.wsdl'); ============================== I was also supplied this HTML: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> When I comment out the PHP code, the page displays. When I only remove/comment the require_once() code, PHP spits out an error at the line that tries to create the new object. Obviously, the folder I've specified is incorrect, because I haven't installed anything myself - but where is the SOAP package really, and how do I access it? Or, is the class automatically known, and I'm doing something else wrong? Thanks, all. Dorsey
  9. I should have mentioned that there are almost 50,000 ZIP codes in the U.S., possessions, and territories. The algorithm to locate ZIPs within a radius of a given ZIP Code is five lines of PHP for the math and one select. Incredibly simple. Dorsey
  10. OK. About two hours after posting this, I found what I was looking for, and have implemented a decent locator. The most informative seems to be: www.greatdata.com. Of course, they want to sell you their database and software, but there's a free demo and an explanation of how to use it. They even accomodate changing latitude in their longitude calculation. Another source for info is the U.S. Census Bureau at www.census.gov/geo/www/gazeteer/places2k.html. These folks only have info from the last U.S. census (2000), but it's extensive and free. The USPS also offers ZIP Code products for $25. Do a vivisimo search (or Google, if you must) on "zip code locators" and you'll find a lot of stuff. Dorsey
  11. I'm looking for something, either software or an algorithm, that would allow me to determine which ZIP codes surround a given ZIP code within a mileage range. I know that this is a somewhat nebulous request, but I'm looking for similar capability to map sites that locate businesses within X miles of one's home, given X and your home's ZIP code. I would hope this might be easier than implementing a map database and working at the street address level. I suppose that I could obtain geographic data with map coordinates (latitude, longitude) of all known ZIP codes, and do the math. That may end up being the answer. Thanks, all. Dorsey
  12. How do you specify the actual file name? I'm trying to use a simple counter file, and have had no success opening the file with: $fp = fopen( $counterFile, "r+" ) or die( "Can't open counter file [$counterFile]" ); It fails everytime. I know the file exists, although I can't seem to change the permissions from 644, but that's another issue. I've also tried "r", "w", "a", and "a+" options with the exact same result. Do I have to give the full path? I've tried just "counter.txt", "www.ebackbid.com/counter.txt", "/home/ebackbid/public_html/counter.txt" and "public/counter.txt". I'm running out of permutations (and patience). Thanks, Dorsey
  13. I understand that dial-up lines could create an artificially high visitor count, but what about AOL users who all appear to come from the same IP producing an artifically lower visitor count? We're trying to guage the success of our marketing efforts, and REALLY need reasonably accurate visitor counts. Dorsey www.ebackbid.com
  14. It's probably a start at what I'm looking for. If you don't mind, please send me a link, or show it in this forum. The script I wish to "profile" consists of outer and inner query loops. Within the inner loop, e-mail is generated, but not necessarily on each iteration. It would not be uncommon for the innermost code to execute millions of times. Other RDBMS will show their query execution plans for given SQL, but I'm not aware of this facility with MySQL. Evaluating a QEP can reveal how effective indices, key selections, and table organization actually are. Honestly, I don't have formal MySQL DBA training, so there may be something I'm not aware of. So, it's really this that I wish to evaluate, and any clues as to how to accomplish that will be appreciated. Thanks.
  15. Is there any way to determine the execution times of scripts? I'd really like to "profile" my script before it gets out of control. In the past, I've used run-time profilers to determine where and how much time is spent in programs I've written, with the objective of determining if new algorithms are required, or if re-arranging the code can reduce runtime and/or lower response time. I've looked into the MySQL and PHP documents, and their respective forums, as well as TCH forums, but haven't found anything like what I want. These scripts will be run as cron jobs - is there something available along those lines to obtain the information I need? Or, is this whole concept a thing of the past? Dorsey
  16. I'd like to be able to test PHP scripts before committing them to cron jobs. Is there a way to do this? In a past life, I used telnet or the equivalent to get a "terminal window" and prompt. Thanks for your help. Dorsey
  17. I'm looking for an alternative to VeriSign. While I've used them in the past, it doesn't hurt to have alternatives, so I'm asking folks here what they've used, and their experiences with the various products. What I really need is something analogous to Payflow Pro, which is a PHP-callable interface, as opposed to Payflow Link, which is a callback mechanism. For the typical online sales situation, where the user puts stuff in a cart, then moves to checkout, Payflow Link is acceptable, but I now must perform batch billing of accumulated charges, and Payflow Link is not feasible for that purpose. Payflow pro seems like a good solution, but the library must be purchased and built into Apache, and I'm not ready to step up to a dedicated server at this time.
  18. Dorsey


    Thanks for this information. I picked up the tool and the instructions, and it's working just fine. I would recommend this to anyone.
  19. Dorsey


    How do I send (I think) an icon so that something other than the IE or Netscape icon appears in the Favorites list? I've seen this several times, and don't see how it's done by looking at the source code of sites who do it. If this is the incorrect forum for this, please re-direct me to the correct one. Thanks.
  20. I've followed the instructions in both the MX documentation and others in this forum when trying to upload files to my domain. It consistenly fails, so I also started checking the FTP log (duh) to see what was sticking in it's (MX's) throat. After validating the login and password and successfully changing to my root directory, MX tries to issue a PORT command (PORT 192,168,1,3,4,117). Every time, this hangs the connection, which subsequently times out. Having used FTP manually for years, I don't recall ever having to issue this particular command. What does it do, and is there something that has to be enabled/set up on either side (MX or TCH) to allow it to work? Thanks for your time. Dorsey dharrington@erols.com
  • Create New...