Does anybody uses Realm Cloud for production?

cloud

(Max) #32

@Sipe A bit late answer, but yes, we need offline sync.

My (hopefully) last 2 cents to this thread. Continuing @valokafor thought:
Yes, 350 paying developers out of 100,000 is almost nothing.
I don’t mind to pay but a reasonable price.
I do not see much difference between $30 and $250 plan for me. The second one does not provide 8 times more services or resources. $250 plan still offers only 5x9 support.
Dedicated plans does not scale well and

The most reasonable pricing model would be ‘pay-as-you-go’ or pay for consumed resources.
Montly Active Device means that I should pay equally for the user who is actively consuming resources and the user who has just tried application once and uninstalled it.

By the way, I hope if we started to discuss Firebase it is not prohibited here to mention other products.
Does anybody give a try to Azure Mobile Apps? The pricing is complicated there but at least it should be reasonably scalable.

P.S. I have left a comment under MongoDB SEO post here https://www.mongodb.com/blog/post/mongodb-and-realm-make-it-easy-to-work-with-data-together though he didn’t noticed it :grinning:


(Max) #33

Answering myself: Azure Mobile Apps are mostly dead.


(Robert Oberhofer) #34

I want to do a last comment to close this thread as well. I am on the Product side.

Your comments here have not gone unnoticed.
We agree that usage-based pricing does make more sense and are discussing how (and even more so when) to enable that.
This will be possible as we better integrate Realm with the capabilities that MongoDB provides (especially on the Atlas side). There is some work involved with that before this can be rolled out.

We believe that the resources available for the $30 account are pretty generous, thus you can run smaller scale applications on this tier. For purely development a lower tier would be beneficial. Something we are looking at as well (prior to rolling out usage based pricing).

We (and this includes MongoDB) are committed to make Sync a prime way for client/cloud communication. Much easier with the additional resources that MongoDB is putting into the project. Plus we now have additional resources to improve the Realm Client DB as well and address some of the outstanding issues faster. This should be good news for everyone.


(Max) #35

@robertherber
Thanks for your reply and explanation of the situation.
As far as I know from your Sales department you have stopped selling Dedicated plans at all until MongoDB decide on future licenses and prices.

  1. What are your recommendations for those who overgrow $30/month plan and need more reliability and resources to move to production?
  2. Are there any estimations regarding when the new pricing models will take effects or any decisions are taken? At least, approximately (1 month, 3 months, 6 months)?

(Realm) #36

@wildfisher:
To 2:

  • The estimate is that the integrated Realm/Stitch MongoDB product will take us into Spring/Summer 2020.
  • With the integrated roll-out we will support usage-based pricing, since this is the default model for MongoDB Atlas. Anyone who (might have) used Atlas will be familiar with it.

To 1:

  • In the meantime if you outgrow the $30/month plan then please contact [email protected].
  • We have to work out a plan to support your case. If you reach that limit you might want a dedicated instance for your app anyway.

(Max) #37

I have contacted sales but they say they know nothing and stopped selling dedicated plans until receiving new instructions from MongoDB.

My last question to support concerning technical problems was unanswered for a month and now the support silently closed the ticket.

It’s all very sad.
It looks like ‘sometimes in the future everything will fine’ but for now there are no chances to take it seriously and rely on Realm for any public application.


#38

@wildfisher

Apologies for the unanswered technical issue in the support portal. As Nikola and Robert mentioned before, we make concerted attempts to support our community Sync users (Standard, $30/month plan) with a best effort, but must create a balance with supporting our enterprise deployments.

What is the ticket ID? I will look into this for you.


#39

How do enterprise deployments handle sync errors in a layered architecture? This question has had no response from Realm and so reinforces the concern we have about Realm Cloud being ready for production.

I really do like Realm and would be delighted if these concerns could be allayed.


(Nikola Irinchev) #40

I’d like to reiterate a few things already mentioned in the thread because it feels like we’re going in circles.

If you have an app in production (even on the $30 plan), we’re going to do everything reasonably possible to make sure it’s working as intended. If that means moving you to a dedicated plan, we can do that. We’re trying not to sign up new customers that would need a dedicated plan until we’re further ahead with the integration with MongoDB’s infrastructure, but that doesn’t mean we’ll just abandon existing customers who may have apps in production.

There are enterprise/sophisticated apps in production running on Realm Cloud. The forums is a place for community support/discussions and we’re not forcing any of our customers to sign up or discuss their use cases with others. We have a best effort commitment to replying on questions, but the forum will always be lower priority than our support tickets. So if you have an urgent issue/bug report, file a ticket. If you have general questions around architecture and have a support contract covering code reviews, then file a ticket. If you have neither of those, we’ll try to reply in the forums as time permits, but you shouldn’t expect 8 hour response times or responses tailored precisely to your use case.


#41

That’s what happens with forums though … there are threads on Stackoverflow for example that run on for years, meandering all over the place and as long as everyone is respectful they can be jolly good fun.

But more seriously, this particular thread asks a question that remains pertinent. I really like Realm and I want it to succeed as we think it has great potential for the markets we develop for. But we are still doing due diligence on Realm.

We aren’t confident we can move to GA with a Realm product yet and we have a responsibility to our users, small businesses, who rely on our software to run their daily business, pay the rent and feed their families. We take seriously our responsibility to those who have put their trust in us over the years.

We have raised tickets in the past. One of our contractors raised a ticket as a non-paying user and was told that priority goes to paying users. Fair enough, but when we have raised tickets as paying users and the response is not as prompt as we had hoped, we are told we should upgrade our plan.

One example is where one of our customers had a corrupted realm on Realm Cloud. We raised a ticket the day he notified us, but Realm support took a week to roll back his database. That poor guy earned no money for a week!

With a response like that is it any wonder users raise issues publicly in this forum rather than raising a ticket privately?

Also Realm support disappeared for a week. How could something so important and so simple take so long to fix? That sort of thing rocks market confidence and we quietly wonder if Realm is going to be able to hang on, MongoDB takeover notwithstanding. Speaking as a paid up Friend of Realm, I hope so.


(Nikola Irinchev) #42

Those are fair criticisms and admittedly, we have had to stretch our support resources thin to make sure our enterprise customers are covered, which meant that people on the $30 plan were suffering longer-than-what-we’d-like response times. That’s expected to improve as we integrate with the larger MongoDB support organization.

Similarly, things have been a bit hectic while merging the teams, so fixing the support portal kind of fell between the cracks. But again, after this initial transition period, we expect things to start working much smoother.

Finally, if you don’t feel the product or the support team are just there yet for your use case, that is your call and no one can argue with that. I believe we’ve come a long way in terms of stability over the past year and, I expect things will get even better with the MongoDB acquisition. That being said, if you have a project that’s running into deadlines and are worried Realm will slow you down or prevent you from shipping a polished product, then perhaps you should look for a different solution. I do hope that in 6-12 months, you’d be willing to give it another shot, though :slight_smile:


#43

Thanks for the candid response @nirinchev. It has looked like Realm staff have been missing out on sleep.

We don’t currently rely on Realm, but we will keep a Realm version of an app available in prerelease for our users and hopefully we’ll see things settle down. I hope to come back to this thread at some point in the future and post that we are successfully running Realm in production.


#44

Update: I’ve used realm in production for some time now. I’ve had a long stretch of near to no problems. But I think a recent sequence of Realm Cloud updates has been the cause of some problems that I’m now experiencing. Luckily, I’m not actually charging for the product.

Realm as a local DB is great, but as a platform it currently feels a bit frail. I am beginning to doubt that it is something for me.


(Robson Barreto dos Santos) #45

I am using Realm Cloud for production with the standard plan ($30) for something like 2 months now. In this period my instance went offline 2 times, and i have some good and bad impressions about these episodes:

Bad impressions:

Support takes too long to answer critical issues. When my instance went offline for the last time, I opened a ticket that hasn’t been answered yet (4 days waiting). Fortunately, my instance mysteriously came back;

Mine logged out users weren’t able to log in until my instance return. I miss a login offline feature, that will avoid a lot of troubles for sure;

Good impressions:

Fortunately, my logged in users were able to continue using the local database and sync their data when my instance came back;

And of course, besides my impressions about this dramatical episodes: I certainly endorse that Realm should be more transparent about prices and consumption metrics!

[EDIT]: Actually, @eric.mossman answered my ticket 2 days ago, but just saying that it will take a couple more days to get my ticket.


(Yuval) #46

I’ve been using Realm Cloud (30$ plan) for almost 1.5 years now.
We have several thousands of users .We use synched realms.

Apart from today , my last support ticket was opened in October 2018.Since then until today the Clouds was stable and worked great.
Today, my production server suddenly had an issue with registering new users.
I opened a ticket 8 hours ago, got an answer about 2 hours later that the issue was found and forwarded to the technicals and since then I heard nothing.No update and unfortunately no solution.I have since emailed twice asking for update but nothing.Since I opened the ticket the system seem to be unstable - some users are connected but see no data although it is stored in the cloud.
This affect us badly.Loosing customers ,in simple words.Still waiting for support to come back to me.

So the feeling is that if all is working its working good but if something goes wrong it may take long hours before the issue will be resolved.These were the times also in my last tickets.

Very frustrating.

I read all the posts including the answers from the Realm guys.
I don’t mind paying more to get a better service.I offered this to their sales guy a year+ ago when it took them hours to bring my production back.Unfortunately im sharing the same feeling that paying more will not get me the support and service I need.

Our users are not heavy consumers of data and the app cost is far below 20$ a year so expensive solutions will not work for us (dont need to talk with sales to understand that).

I understand the priority for Realm customers paying the premium plans and agree with that but I think that giving solutions to downtime or for example my case - unable to register or login - should be core support as these are core services .(again - in the last 10 months I didn’t have a down time so not all is bad ).

To sum it up I think the response times and resolution times are unacceptable in basic plans (30 to 250) and a company cant work with that.

Forum is very weak tool.Posts are not always answered and you are left alone.

Pricing ,like some mentioned, is not transparent (I actually only now read about the 5x9 and 24x7 options - not much in the forum as I wrote before).

Im quite concerned myself after reading this thread and thinking if starting with Realm was a mistake as we are deep into it.

On the positive side - when there are no issues its working great.We got used to it and I hope we will not need to find an alternative.

I will be happy to pay more to Realm and know that I get the support I need.The question is whether I will get it when I need it.

So - when all is good - all is good, but when something is wrong it might get messy.

Still hoping that someone will reply my ticket and my users can use the app again…

[EDIT] It’s been 25 hours now since I opened the ticket and 23 (!) hours since I last heard from support .The issue still persists , new users can’t register with the app and existing users experience issues.Unbelievable and disturbing…feeling helpless


(Brian Munkholm) #47

Hi @yuvalkro
We are really happy to hear that you have been using the service for 1.5 years without any major issues apart from the one mentioned, which turned out to be an overload situation on your instance that was resolved during the weekend.

We surely sympathize with the frustration when your app is not working for your users! Nothing worse than that. We feel exactly the same when our service is not working for you…

That said, although we actually try to surpass expectations, the issue occurred over the weekend and the $30 plan doesn’t come with the SLA expectations you seem to need, so I would encourage you to get in touch with support and sales so that we can provide what you need to be successful and happy with the service.


(Yuval) #48

Hi @bmunk

It took 46 hours to resolve the issue and 44 hours to update me.For 44 hours I heard nothing back from you.No update, no status. Things can go bad, we know that as developers , what you would expect is at least an update.

Allow me to paste the reply I sent to Nick who answered my support ticket .
I think it describes best where the frustration is coming from.More than a year ago I suggested your team to pay more and get a better SLA (you didnt have better SLA back then), so its not that im not willing to pay more, I do, I am just wondering whether your next SLA would have handled this issue better.Especially see my pre alert comment.

Yuval

Nick,

“Your instance was still operational and handling requests, but it was taking a long time to process some.”

De facto my instance couldn’t register new users and users didn’t see their data .It wasn’t slow.It was partly not functioning.
If I didn’t hit any of the factors YOU set as limits to the plans why do I need to worry about this ? all what happens behind the scenes is a service up to you.
The contract between us says that if I reach close enough or over the limits you set then I need to change plans.Not because of the structure of my data.Can you point me to where its mentioned?

If the instance on the cloud (AWS if I was told correct) need more resources , and not due to the factors you set as a company for pricing a plan (yes , even the 30$ plan) then the problem is on your side.
Either remodel your cost plan and factors or allocate more resources to the instances.
It is not acceptable to come to a paying user and say that resources have run out while he didn’t cross any limit YOU set.
I had 1/4 of the bandwidth and mayb 10 of 10,000 concurrent users.
I can’t measure any other factor , and if you can you should put this as a limit in your plan and not give it as an excuse that I as a customer have nothing to do with it.

My instance on YOUR servers is performing poorly ? raise a flag.Why didn’t I get an automatic alert ? how can I know im about to face a problem when you don’t give me a heads-up? why I was not contacted before? and why when I did contact it took you 40+ hours to solve this?

And finally, “Best effort” is not an excuse for you guys to perform poorly.Its not the first time I hear about this and I read about this also in the forum.You are acting as an amateur while we, the developers and business owners put our trust on you.Very bad news for Realm.Very bad news for those who rely on you.If you were working correct there was no need to prioritize this issue.An alert would have sent to me and we could have discussed this issue beforehand and avoid me facing a crisis with my customers using my app based on YOUR service.
The excuse of prioritizing and “Best effort” doesn’t do well with you.


(Brian Munkholm) #49

Hi @yuvalkro
It wasn’t my intent to discuss your particular issue in details in this thread. That will overload this thread too much I think. We can do that in the ticket itself. But surely, we fully agree that other operational limits should both be known and at least alerted, and we are striving to improve that area of our monitoring/scaling. Most of our services scales automatically, but there are still a few areas, in particular for the $30 plan that doesn’t do that yet. Normally very few of our $30 users hit any limits, so we are still learning what those are to take preventive actions on them.


#50

We’ve had a Realm Cloud app in pre-release for a year or so. We’ve had a few scares including a couple that caused data loss for a couple of users. Not a lot of data and not a lot of users (we have about 150 users on the pre-release) but enough to keep us awake at night worrying. Enough for us to warn users that we are unable to guarantee the integrity of their data.

One user who experienced data loss was out of business for a week while we waited for his cloud database to be restored by Realm (Ticket 3916).

Recently our Realm Cloud server fell over and rebooted repeatedly. We raised a ticket marked urgent (Ticket 4536). To be fair the initial response to say they were looking into it was quite prompt. Then nothing for a week. In the meantime the server righted itself but hundreds of partial realms had been created. The support person noted the server seemed to have become stable and closed the ticket and raised another regarding the bogus realms (Ticket 4574). I’m left with the impression nobody at Realm knows what happened.

It’s two weeks since Ticket 4574 was raised. Nine days ago we were told that due to a support backlog it might be a while before they get round to looking at why Realm Cloud created hundreds of bogus realms on our server. So far we’ve waited almost a month (across two support tickets) for an answer.

Which brings me to another issue. Realm have decided to rebrand paying subscribers like us paying $30 per month as “Community Users”. That really doesn’t wash. We subscribe to a number of web and cloud services to support the varous products we offer. They are all typically around $10 per month or less, yet they consider us as subsribers, not “Community Users”. I susect this rebranding of paying subscribers as community users is an excuse to cover terrible support.

Realm support has deteriorated dramatically since the acquisition by MongoDB. We are left to wonder if this is due to the conditions of the sale. Have Realm been forced to all-but-abandon support in order to meet MongoDB imposed deadlines?

I want Realm to succeed as it solves a number of problems for our small business users. In posting our experiences I am aware I may be hammering another nail into Realm’s coffin. However I feel uncomfortable about not sharing our experiences with other developers considering committing to Realm Cloud.


(Kurt Libby) #51

We have been using Realm with sync in production for a couple of years. We recently moved from self-hosted to the Cloud. We have approximately 200 users that have migrated to our cloud instance in the last week.

I woke up this morning and was attempting to access an account and it wouldn’t log in. I checked our instance and it currently says that it is “Getting Ready” even though it’s been used in development for about a year and in production for a week. We have another instance that we are using in our development environments and it is running fine. I did submit a ticket and then came to check the forums. After reading these 50 posts, I’m a little nervous that this will go on for awhile. Hoping that is not the case.

But to answer the question, yes we are using it in production. It was ok for a week. We’ll see how this goes.