Hello everyone! Today I will be addressing incredibly important matters, starting with breaking down the complications that came out of the migration, and how we are going to clean up the mess it left us in. I will also be sharing the big announcement about the changes we are making, and how your support will play a hand in the blog having a future. For that reason, I urge you to please take the time to read over this.
To help you understand how we ended up in this situation in the first place, I will be sharing you the full story of the tireless work that was being done behind the scenes. The past four months has been an incredibly chaotic series of events. To say I am emotionally and mentally burned out is an understatement to say the least!
Let’s begin by going back to 2017/2018 when I first embarked on this quest of insanity.
The Time Bomb
The migration to a new server wasn’t something I decided a few months ago, or last year. But rather something I had been planning and tirelessly working on ever since the big crash that transpired in Fall 2017 – Winter 2018. For those who had not been with since us, it was a period where the blog was more or less on life support to due a fatal flaw in the server, making it virtually inaccessible and incredibly slow. While we were able to get some help and eventually resolve the problem, I understood this was only a bandaid solution to buy us time. For how long? I had no idea. As result, the fear of the blog experiencing that kind of incident again was something that has been hanging over my head ever since.
With no time to waste, I spent a lot of time on research, searching up the criteria we needed to avoid encountering such problem again. But much to my dismay all of our options seemed to be out of my budget. Nevertheless, I stubbornly persisted on and eventually I learned about the new hottest commodity that would better accommodate our needs, VPS aka: Virtual Private Servers. It was something I had learned a bit about in my research efforts, but it was admittedly a daunting idea for someone as myself who knows nothing about that kind of thing. When it comes to programming, coding or anything related to the starts, I know a little bit of the basics, but nothing more than that. But the problem with VPS, they can be extremely expensive and difficult to manage on your own. Dissuaded by the price tag and complexities involved, I put it aside and continued on with my search.
APRIL 2018
Summer 2018, Oki refers me to someone to work with. We work together and settled on a plan. However due to life hurling various complications interfering with the progress, we weren’t able to move forward we had hoped. After much deliberation, in October 2019 I decided it would be in both of our best interests to called it off and look for someone else to work with.
OCTOBER 2019
Fall was an anxious period. First week of the Fall Season was botched by the site experiencing downtime and at the same time, I had noticed it had gotten increasingly more sluggish. Alarm bells were ringing in my head. I felt a sense of deja vu of the 2017 Crash. It reignited the urgency to ensure we move to a brand new server for Spring Season. And with quite frankly, I didn’t have a lot of faith it was going to last long enough until then. Desperate to find someone to work with, I spoke to everyone I knew with backgrounds on I.T., computers and programming in hopes they maybe knew someone I could get in touch with.
FEBUARARY 2020
It wasn’t until five months later, I seem to catch a lucky break when I was referred to another person to work with in hopes to help me finish where I had left off in the last process. However as we discussed how to proceed, I learned about a number of things from the fact it was not something I could manage on my own and the cost was actually out of my budget. So I scrapped the plan that had been worked on for the past year and half, and reviewed my options once more. But as they gave me the run down and it was just as had initially feared from the start: Migrating our all of our content to a service that would better accommodate us was completely out of my budget.
It was then when I came to accept that no matter what service I choose, or what I do, the hard truth is that it is no longer financially sustainable for me to shoulder the costs of running the blog on my own.
So I called about a Team Meeting to get everyone up to speed on the situation. I presented them our options, and after much discussion we have reached an unanimous agreement it would be in the blog’s best interest to take the most expensive option that will give more stability and lots of room to grow in the foreseen future. To do so, we all agreed from this moment on, every writer on the team would be required to chip in to help me cover the costs for the new server.
Now that the costs were partially covered to ease my financial burden, it seemed like things were finally on track to move forward with the big move. To ensure we wouldn’t be unnecessarily billed for an extra month, we decided to begin the process that was said to only take a couple of days to complete in the beginning of March.
MARCH 2020
It is time to finally get the ball rolling, and on the evening of March 3rd, I am given the total cost for to set up the server and migrate our content.
The moment my eyes landed on the price, and it triggered a panic attack.
I couldn’t believe it. I had thought my eyes were fooling me. I couldn’t understand it either.
I had made it crystal clear from the very beginning about what my budget and limitations were. I have no idea where this guy thought I was going to pull the money from!
Now having reached a dead end and with little time to spare, I declined to move forward and hastily began scrambling for a Plan C. It was this particular event was what forced us to post-pone the big announcement that was originally scheduled to be released sometime between late February and the beginning of March.
Six weeks until Spring 2020 starts, and here I was having a mental breakdown because our older server was increasingly becoming more and more unstable with more downtimes and taking even longer to load. I was terrified that we were running out of time.
Then, not even 24 hours later, the unthinkable happens. Quietcupcake refers me to a friend who heard about our situation and offered to lend a hand. And with their help, we were able to quickly come up with not only a more economical solution, but a pragmatic one that would better equip us for our future growth.
For starters, what we needed to do was cut down the blog size. To give you context, between 2010 – March 2019, it was 260GB. This is why our OG provider was telling us we were too big for their service and recommended it was time for an upgrade. It was taking a tremendous toll on our server, on top of the other technical problems we had been experiencing over the years. To do that, we looked around and settled on a Cloud Service that was within my budget. Next we needed a plugin to serve as a bridge between our blog and the cloud. By the time of that was settled it was already the middle of the month, and with our looming deadline for our server renewal approaching, I spent the next two weeks staying up all night to complete the process.
Since we couldn’t just do a full bulk import, I had no choice but manually select all the images each month from March 2020 dating back to September 2010. To import to the cloud, the average time it took to do each month was 2 Hours. So that was 226 Hours of work to do in 2 weeks time, all while juggling multiple things such as, keeping things organized with the team, finishing up the Spring Preview, backing up all of our content, preparing for the server migration, writing up my entries, and real life responsibilities I have to take care of.
Somehow I made it. The last stretch required 3 sleepless nights.
Where it all Went Wrong
Finally it’s almost D-Day, it’s time for the migration. The new server has been all set up and we are ready to move! Everything was smooth sales… until the migration process happened. It was only then we encountered a series of unexpected problems…
#1 – First import attempt partially failed + Broken Categories
It imported all the users, categories, pages, and other stuff, but not the posts. Shortly after that I discover the import broke the categories, and I have to re-appoint every category to its parents. Since posts were missing, I began importing the users one by one. And then the second issue came up…
#2 – Suddenly WordPress refused to import Users, claiming the file didn’t exist.
I knew that was bullshit and it had to be bugging out because it was working fine up until that point. This is why we only have our active writers’ content uploaded to this server, and some of our categories are left blank. It took about week before it finally sorted itself out, however, before that, yet another problem popped up…
#3 – The Cloud’s Bridge Broke
I am really salty about this one. After being forced to pay for another toll for the Plugin I was using to bridge the image content between the blog and the external cloud, just after I successfully finished importing JAN 2020, FEB 2020 AND MARCH 2020, IT STOPPED WORKING.
But that wasn’t all…
#4 – All Galleries and Feature Images were All Wiped Out from Entries
This is probably the biggest travesty of the whole thing. Not only did I waste money on this (and I am already going over my budget as it is just to resolve the problems), but now after working 226 Hours for the past two weeks with little to no sleep, in the end, it was all for nothing.
#5: SSL Set-Up Problems
Considering I’m a total noob when it comes to handling domains separately since I have always been using the packaged deal with shared-servers, this was something that slipped my mind as I thought the process was going to be straight-forward and simple (obviously it was not).
Of all the problems, this was definitely the second biggest headache of all. For four days, even though the blog was live, most of you probably couldn’t access it. So as long as you were using the URL with https, you would encounter the blank page of doom where you couldn’t connect to the site. It was particularly stressful because I was worried about people thinking our blog was good as dead. Ultimately the solution to this problem ended up coming with an additional cost, but at least it was a one-click solution that relieved me much of the stress that was eating me up inside.
This event with the website not being fully connected to everyone highlights the importance why we strongly encourage you to follow or at least bookmark our social media platforms such as twitter or tumblr since they are the only places we can make any announcements outside of the site in times when we experience technical problems. (But now that we moved, much of that should resolved now! But still, please follow us, this is the best way to keep up to date!)
Going Forward
Once the SSL issue was resolved, for the first time in months, it allowed me to take a step back, take a deep breath and get some much needed sleep for a few days. Now that I wasn’t as sleep deprived as I was before, I took the time to assess the situation with renewed focus. There were a lot of things that were messed up in this migration, some of which I hadn’t noticed right away (such as the broken categories), while some other things, like the migrating our former writers, took some more time before it could be resolved. At the time same, I have also discovered some of the ways we were going to do things, was not going to work as we hoped.
This is particular problem with the whole plan behind re-importing our content images from our cloud, back onto our website.
As the team and I were working on restoring the galleries and featured images from the media content I was able to import JAN 2020 – MARCH 2020 content before the plugin broke (preventing us from importing content from 2019 and older), we discovered the process of navigating through the media library to find the content was incredibly inconvenient. The first problem was that the images were no longer divided by the months they were originally imported in. So instead the content that was originally released in say June 2018, because it was re-imported from the cloud in March, it would be considered new March 2020 content. That was when I realized, even if the plugin’s problems were to be resolved, the more months I upload, the more cumbersome it was going to be to search for the images in the media library.
The other problem with this is that, regardless if the writer had named their file properly or not, the media library’s search engine doesn’t exactly work the way you would hope it to. This makes it a plainly tedious process for those who did it correctly, where as for others, this would quickly become akin to looking for a needle in a haystack.
And finally, my biggest gripe with re-importing the images is that fact it would take up to 2 Hours to do every month all over again. That alone takes up too much time. I am not about pull another two weeks without a wink of sleep just to get it done. Not to mention, I have no interest in renewing the subscription for a broken plugin.
So we ultimately decided we are going to directly re-upload all the images to its respective posts. For some writers who no longer have their screenshots will be required to redo them. For others who actually held onto them, this process will be a much swifter.
So here’s how we are going to restore all of the lost Snapshots, Galleries and Featured Images…
THE MEDIA RESTORATION STRATEGY
Step #1: Re-Organizing the Categories
Since the Categories are all messed up, I need to re-sort them back to their parent categories.
We are also going to use this opportunity to properly establish rules for the categories as well.
- Completed: Unchanged
- Currently Covering: Unchanged
- Dropped: Series that were dropped at/or after Episode 5 (regardless of reason)
- Quick Look: Now only applies to series that were covered for four episodes or less.
I am going to try my best to have this part fixed up before we are set to switch over, everything will be easier to navigate again.
Step #2: The Pragmatic Strategy to Restoring Media Content
To be more efficient with our time, for the series that are in the categories of Dropped or Quick Look, I suggested we skip the galleries altogether, and just slap on the featured image, and/or drop a couple of Snapshot images (2 – 4). This will help speed up the process.
Next, for the time being: We are aiming to restore content dating back as far as 2015. The restoration of content older than that will vary on its popularity and will for the writer to do so.
As for the pace work at, that too will be determined by the writer. I understand how much of an undertaking this is going to be, so it is very important to us to take measures to avoid burn-outs.
To get an idea of how long this process could take, here is purely theoretically timeline.
Let me emphasize the team is not expected to follow this pace. I pulled this out of a hat.
Theoretical Timeline Estimates
Writers with less than 100 Entires
- 7 Entries/Week – 1 Entry/Day = 13 Weeks to Restore
- 14 Entries/Week – 2 Entries/Day = 6 Weeks to Restore
- 21 Entries/Week – 3 Entries/Day = 4 Weeks to Restore
Writers with less than 150 Entires
- 7 Entries/Week – 1 Entry/Day = 21 Weeks to Restore
- 14 Entries/Week – 2 Entries/Day = 10 Weeks to Restore
- 21 Entries/Week – 3 Entries/Day = 7 Weeks to Restore
Writers with less than 350 Entires
- 7 Entries/Week – 1 Entry/Day = 50 Weeks to Restore
- 14 Entries/Week – 2 Entries/Day = 25 Weeks to Restore
- 21 Entries/Week – 3 Entries/Day = 16 Weeks to Restore
Writers with over 600+ Entries
Focus on Restoring content all the way back to 2015 (going back further is optional), but prioritize series based on its popularity and demand.
OPTIONAL – Step #3: Rinse and Repeat, but with Former Writers’ Entries
Once we finish restoring our own respective entries, we will begin restoration process of our former writers’ content. This will be approached a bit differently.
As I understand this monumental task and that this isn’t anyone’s responsibility, this task will be one writers can choose to opt in if they so desire. To keep things organized, the volunteers will be assigned a former writer they need to gather the featured images from. The pace which those entries are restored will depend on the individual, and whether they have the time, energy and whether they have the will to finish the job.
Because of the sheer amount of entries (their articles, makes up half of our content), I decided the most sensible strategy would be to just focus on restoring the posts’ Featured Image because at the very least because we can just pull it from our social media, such as Twitter.
We will try out best, but there will be some that will have to be left as they are.
Reflection
I will admit, looking back on all of this makes me feel ashamed and embarrassed of myself. There were a number of things that I could have done differently, and maybe it wouldn’t have turned out the way it did. One of the things that might have been better was to have not purged the image content on our OG server once we migrated them to the cloud. Of course I have a backup of their files, we ultimately decided it was a sacrifice that had to be made in order to properly grasp the true blog size and stabilize the server that was not looking to hot at the time. The blunder with the domain name and the HTTPS being a cockblocker because I didn’t have it properly installed at launch is another mistake that could be been avoided had I been better prepared. And then I thought I had a handle on things, I faced new were problems I have never encountered in my 10 years of running this blog.
But there is no use of dwelling on what I could’ve, would’ve, should’ve done. I also recognize I did the best I could with the incredible time constraints we were working with. It wasn’t easy, but I got through it, and I learned a lot from this experience. I should also consider myself lucky that it wasn’t an even bigger mess, and that while it is going to be long and hard job, at least it can be fixed.
TLDR: Even veteran bloggers like myself make dumbass mistakes. If you can, don’t rush things, and most importantly: Sleep, or else you will go insane.
Support AngryAnime Bitches Anime Blog
For much of the blog’s lifetime, it has been a privilege of mine to be able to able to sustain the server host costs for many years on my own. And during the times the few times I was not in financial position to pay for the renewal of the server, I have been incredibly lucky to have received generous donations from our writers and readers like you. It is thanks to your contributions during those hard times that I was able to continue to provide this platform for 10 years!
And yet, even after receiving your support at the times of need, I have always been reluctant about asking you and our writers to chip in what they can, or even leave the donation tab up. This is because I created this blog as a space to build a community for both our readers and writers to freely express themselves and their passion for anime. But I understand now more than ever that I need to open myself up to a lending hand. It is no longer financially sustainable for me to shoulder the costs on my own. And it is now especially true since the COVID-19 pandemic put me out of work due to the government’s shutdowns of businesses and orders to quarantine ourselves to plank the curve the best we can. It is said these orders could last as long as till the end of the summer, but at that point, I don’t even know if I will have a job to return to. As such, this crisis further highlight that in order to to keep this platform running, I need both the team and your support.
After discussing the matters with the team, we have reached an unanimous agreement that as of February 2020, all writers on our team (including part-writers and those on hiatus) will be required to pay a flat membership fee to help cover the costs. Everyone, including our recently retired member Midnight readily chipped in. So please let me express my gratitude to our amazing team, who despite everything being last minute were able to immediately pitch in so we can afford to make the move as soon as we can! Ideally I would have liked to have given everyone a heads up in a year in advance, so truly, thank you from the bottom of my heart. It took a lot of weight off my shoulders.
So now I would like to ask of you, our dear readers for your support to help ease the financial burden for the years to come. You can do so by becoming one of our Patreons!
At long last we have created a Patreon! 100% of the proceeds will go towards the server costs.
Patreon Tiers
At this time, we currently have two tiers.
If this takes off, we hope to be able to add more fun tiers in the future
- Tier 1: Dango Discord Access
In Tier 1, we invite you to our brand new discord community! We hope this will better enable you top connect with our writers and fellow readers who are just as passionate about the series as you are!
- Tier 2: Takayaki: Discord Access + Special Shoutouts
In Tier 2, you not only get our access to our discord community, but we will give you a shoutout in our Seasonal Previews and Patreons’ Section on our Blog’s Sidebar!
If you are able to so, please considering supporting us so we can continue for the years to come.
Thank you for all the support and love you have given to us all these years. We really appreciate it.
A Moment of Acknowledgements
I would like to take a moment to thank everyone involved in the process for helping me through this. It is an understatement to say these past 6 months has been an overwhelming emotional rollercoaster. Thank you to the team for their support and being so understanding of my distress. Thank you to Oki’s husband who had helped us salvage the site in 2018. Thank you to Quietcupcake’s roommate for coming in clutch for the team and teaching me so much in the process. Both of these guys gave up much of their free time to offer us their incredible wisdom and service. I can’t enough emphasize enough how grateful I am to have received their support.
Sincerely,
Eva
