It has been a few days already since the HF. I've slowly migrated my applications to the new library. But there are still some glitches here and there and inconsistencies. Especially something to do with the APIs. The first thing that I had to do was to move to the latest version of
hive-js in all my applications. Recently there have been a few fixes about to support the new HF but the problem for me is that they are not documented anywhere. Even the dhive documentation is not updated to reflect the latest changes.
I guess the readme file of these libraries is the first place where people can go and check about the changes. It is hard to review all the code changes in the library and align accordingly. I should say that the repository maintainers are doing a great job in updating the repo but it will also be very helpful if they update the Readme.md file whenever a new update is happening.
For the past two days I have been fighting this particular issue where I keep getting
RCP Error: Invalid Parameters error messages in my application whenever I do a call to get post details and
RPCError: Internal Error and
RPCError: Request Timeout error message when I try to get account information. I initially thought the issue was only because I was using the old version of the library. But later after having a few discussions with other devs, I found out that the issue was because of a delay between the API and hivemind. Sometimes in the nodes, we are using takes times to sync up with the hivemind.
@rishi556 gave a suggestion to add a 30 seconds delay in my code to solve the issue. After adding the delay I noticed that the problem was not completely solved but the error frequency has reduced a bit. There was also a recent fix made to correct some issues with the failover feature in
dhive library. The fix was applied by @therealwolf. Even after upgrading to the latest version of the library, I still have issues with failover not getting triggered when there are timeout issues. This doesn't happen all the time but once in a while it happens.
There are a few people like @mahadiyari who run their own hivemind instance. Maybe they are able to remove latency and get rid of these issues. But for the developers who are still dependent on the public nodes and libraries, this is still a challenge. I initially thought it was only me but today I also noticed @buggedout mentioning this issue in HiveDevs discord server. I also see that Peakd is also working on fixing some issues recently. Especially the number of issues they are fixing after the HF has increased a bit. Hiveblocks.com is still not available and the Hive Engine Tribe user interfaces still need a fix.
It would be great if someone helps in sorting out these issues. Maybe after that, we can call this HF as smooth and successful. For my problem, one solution I currently have is to write my own failover code and also to make my application retry the request until I get the values. That may not be a great fix but it will at least help a bit. Before I do that, I'm thinking if there would be a better way to solve this issue.