Developer launched site on old version of Magento, lots of memory problems

[dbanbury]

Long story short, signed contract for our developer to develop a site on Magento a few years ago and they finally launched the site last month. It's still Magento 1.4.2. We host the site internally.

We are getting tons of fatal errors with memory allocation and the site is chewing up memory like no other. The current developer has abandoned the project and we are in the process of finding another developer to come in and fix the issue and of course everyone suggests upgrading to the current version of Magento.

We are absolutely wanting to do that, but I also understand that custom functionality that has been done may also be lost and things will need to be reconfigured, essentially rebuilding the site and spending most of our money on our current devs for nothing. 

I wanted to ask if anyone else has been in a similar situation as ours and what their experience has been? We want the transition to a new developer to go as smoothly as possible and also as efficiently as possible so any shared experiences will certainly help us be wary of making any hasty decisions and going down this road again. Was there a lot of functionality loss that needed to be reconfigured? Did you host your site on your own server or in a hosting environment off-site? Thanks.

[dpalme]

You sent how many years having the site develpped? We are new to Magento, so my opinion is shaded not by my experience with Magento itself but IT in general.

Before engaging the services of a new developer, take the time to do the preliminaries and check them out.  Pull a D&B report, which can telll you alot about their own financial health, get references and talk to each one.  Make up a list of questions you have, problems you have encountered and issues you potentially see  and ask tthem all the questions, on the phone or better, in person.  Do ** NOT ** conduct a reference check by email.  You can't hear tone or see facial expressions in an email and trust me, that cringe you see on their face is worth 10,000 email messages and still conveys a helluva lot more than you'll ever get in an email.

There is ** NOTHING ** wrong with hosting your own site, provided you have the skills to maintain the box and keep it updates, including all security patches.  If you don't have that skill set, your better off going with a managed server with someone like Constant or another provider.  You can even colocate your server with a provider so you get the ability to control the box but get the benefit of redundancy.

If it took your developer "years" to get the site up and running, I have to ask the question why someone didn't pull the plug on this project a long time ago.

[dbanbury]

Thanks for the reply. We actually threatened to get our lawyers several times due to the slow nature of their development. Each time they responded with "We're almost there, we're working on this and that and should have answers for you shortly" and so forth. Most of it was just wishful thinking they'd eventually complete it and not trying to deal with the hassle of litigation, which eventually it came to anyway. In any case, there were several months of legal battles in that timeframe as well. It's been a very bad experience and hindsight is always 20/20. We're not going to choose our next developer lightly, I was just more concerned with the work that was already done being rendered obsolete and basically having to start from scratch when we upgrade to the newest Magento, considering the money we've already invested at this point.

[dpalme]

It will depend a lot on how they developed your customization, was it as a sub theme, an entirely new theme altogether or did they modify an existing base them package, in which case you have issues.  That's not to say that you can't overcome them, but before I let anyone touch a single piece - I'd back the entire system up so you always have a reference point.

[chiefair]

Sounds like a flush and restart... Quit throwing good money after bad, it never works out.

Two things kill Magento launches, 1) not paying enough to hire a good developer in the beginning and 2) not paying enough for the resources required to run Magento.

[codinggiant]

I would not waste the investment that you have already spent. Both time and money. I would seek a new developer ASAP.

Having said that, I want to pitch a little defence for their position. Upgrades take time, especially when there is customization involved. With any software, upgrades should be approached carefully. 

I run into this all the time. We started X project on X version of X software. And we get close to launch day and a new version comes out. And the client says, hey shouldn't we upgrade to this new version? And the answer is not always yes. Especially when this will put off the launch, or invalidate hours and hours of testing. You have got to just let go sometimes, and say hey, lets make that part of phase 2.

[RonaldLKouba]

Seems to me that you are indeed in need of a good Magento Developer. There's a saying that describes your situation pretty well:

"If you think hiring a professional is expensive, wait till you hire an amateur."

Good luck with your project and I hope it turns out to be a success .

[Arjen Miedema]

I don't fully agree with codinggiant's arguments. In this case we're talking about a difference between Magento CE 1.4 and the current version Magento CE 1.9. This is a gap of 5 major versions and about 4.5 years since the last release of version 1.4.x.

First of all, if you need that long to develop a Magento webshop, you're not a capable Magento agency or you're even worse at time management. Second of all, if it's an upgrade from 1.8 to 1.9 or even 1.7 to 1.9, I would tell my client to go live with the current version and upgrade during the project optimization in the first couple of months after the release.

[MagenX]

hm, there is nothing to talk about, few years for development is a bit dumb...

when you can just buy a template from themeforest and start your up to date shop in about the same day....

[chiefair]

Totally agree on that one. The money has already been flushed down a rat hole, there's nothing to salvage if it took this long to "Develop" a website. As is indicated, it's shot through with fail and a dump and restart is valid at this point, along with acceptance that the money paid out is "experience gained in how not to select a developer".