How to use EV while its indexing/sorting/updating?

[luckulucki]

Hello,

not sure, if this goes into support or general, but to me, its a "general" problem ;-D

I have big filelists (3GB) and also index remote server. It goes quite well using EV, but it really is a distressful waiting when EV is "Indexing...and building the new folder hierachy...sorting...." and there is nothing I can search for until its done.

Is there ANY way to use EV while its indexing? (I'm not talking about indexing content. I just mean the general indexing of updated filelists or remote servers)

I really don't need the real-time-database and it would be totally ok if -for the time indexing- EV would use the "old" database and build the "new" database in the background.

Indexing takes about 10-20 mins, so in that time, I can not use EV.

Please tell me, there is a hidden switch I haven't found

[void]

Everything is only really designed to index once.

Changing a file list will cause Everything to reindex.

Consider disabling monitoring of your filelists (Tools -> Options -> File Lists -> Uncheck Monitor changes) and manually reindex on some schedule:

Everything.exe /reindex

Run this daily at 4am or some other time where you're not at your PC.



Remote servers will update in the background and will never cause a rebuild.



Disable fast sorts under Tools -> Options -> Indexes to make the reindex faster.

[luckulucki]

Everything is only really designed to index once.
Changing a file list will cause Everything to reindex.

I know, and that's not really my point.
My point is, that its not possible to use EV while indexing. Such a great tool and such a long waiting time.

It "could" be possible that EV indexes in the background (to a temp-DB) while letting the user continue their normal work. Only when everything is fully indexed would the new database be activated, resulting in minimal delay for the user. (maybe it'll use twice as much resources for the time indexing, but nowadays, in my opinion, that would be a small price to pay compared to the user-friendliness it offers.)

But I see, that this is a general change and thats why I wrote here and not in the support forum, I guess

I'll try your suggestions, thank you very much!

Remote servers will update in the background and will never cause a rebuild.

Thats odd, because I read in the statusline, that its indexing the remote server and thats while I can't use the search. But maybe thats simultaneous when the filelists update...

Anyway, reading the Todo-List, I don't see this coming soon, and thats fine with me. Is is always a pleasure to use your program, because it is soooooo useful and spares a lot of time for manual reseach! It has so many outstanding features, so the waiting time for the reindexing is just a small mosquito bite

Thank you!

[luckulucki]

Just wanted to add an image to show that EV is indexing a remote server and its not possible to use the search meanwhile.

2024-08-02 13-14-24_test - Everything (1.5a) 1.5.0.1378a (x64).png (18.4 KiB) Viewed 1104 times

[void]

Everything should be reindexing from memory and not going to your remote share.

This should only take a few seconds.

Could you please send your Tools -> Debug -> Statistics to support@voidtools.com

This will show the last rebuild reason and more information such as how many files are on your remote share.

[void]

Thank you for the information.

Everything is reindexing your Everything Server as too many changes occurred.



Could you please try increasing the journal size on the PC running the Everything Server:

• On the PC running the Everything Server, In Everything, from the Tools menu, click Options.
• Click the Indexes tab on the left.
• Set the Journal Maximum size to: 262144 KB.
  (Increase this to 1048576 KB if you are making millions of changes / day)
• Click OK.

If you find Everything rebuilding again, could you please:

• Wait for Everything to rebuild.
• Check your Tools -> Debug -> Statistics -> Last Rebuild Reason
• What is shown?

[luckulucki]

with
[*]Set the Journal Maximum size to: 262144 KB.
(because we had only about 30.000 files changed (removed/added/modified) on the remote server the last days)

this is the message:
Last rebuild reason: Out of date network index 192.168.178.250 is available

I can not really say, the indexing (or the duration of indexing) has been reduced, but I will invest more time in observing.

thank you.

[void]

Are you making changes to your index in Everything that is also running the Everything Server?
-A change to your index will also cause clients to reindex your Everything Server.

An index change could be:
Adding or removing a volume.
Modifying an indexed filelist.

[luckulucki]

That makes absolutely sense!

It just is cumbersome to wait several minutes to use EV again while it is indexing. Regardless if indexing locally or from remote server, both indexings force the user to do a coffee break Its bad for the heart.

[Herkules97]

That makes absolutely sense!

It just is cumbersome to wait several minutes to use EV again while it is indexing. Regardless if indexing locally or from remote server, both indexings force the user to do a coffee break Its bad for the heart.

I may be out of my league here as I don't know server stuff, but are you running one instance for two different systems?
I don't know how you do it, but for every physical device I have a separate instance, this separates the search history and such too.
I presume this wouldn't fix the one system causing issues, maybe it's that it thinks it's a slightly different system on some or all restarts so it rebuilds.

At least, if you are running one big one for all systems you have, it would prevent the non-server one(s) from being held back.

[luckulucki]

I worked with filelists a long time. They are really a great feature!

But having to create a filelist on Computer B (which is now the EV "Server") and transfer it to Computer A (which is the computer, I work with) was not ideal.
So I tried the "Server" approach, but I did not understood it as it is indended, I guess.

I was thinking(aka hoping), that the Client (me, on Computer A) is getting real-time data from the Server (Computer B) when asking for stuff.

But I am proven wrong: EV on Computer A indexes the data from Computer B such as it indexes the filelists from Computer B which I was trying to avoid, because it causes too many waiting time

Unfortunately, both ways are having the same issues for me. (letting the user wait until data is proceeded, which can take some endlessly minutes)

So I guess, there is no way to avoid the waiting time, as I understand from this thread. When data needs to be indexed, it needs to be indexed.

I would prefer EV indexing in the background and using the current DB while it is refreshing, but as long as there is no such way, I have extra time to walk around the block and take a break. Not the badest thing IMHO ;-D

[Herkules97]

I worked with filelists a long time. They are really a great feature!

But having to create a filelist on Computer B (which is now the EV "Server") and transfer it to Computer A (which is the computer, I work with) was not ideal.
So I tried the "Server" approach, but I did not understood it as it is indended, I guess.

I was thinking(aka hoping), that the Client (me, on Computer A) is getting real-time data from the Server (Computer B) when asking for stuff.

But I am proven wrong: EV on Computer A indexes the data from Computer B such as it indexes the filelists from Computer B which I was trying to avoid, because it causes too many waiting time

Unfortunately, both ways are having the same issues for me. (letting the user wait until data is proceeded, which can take some endlessly minutes)

So I guess, there is no way to avoid the waiting time, as I understand from this thread. When data needs to be indexed, it needs to be indexed.

I would prefer EV indexing in the background and using the current DB while it is refreshing, but as long as there is no such way, I have extra time to walk around the block and take a break. Not the badest thing IMHO ;-D

I suppose, that and drinking water

Clearly I don't know the rest of this, because I have yet to use two separate systems. Maybe I can look at threads like this if I ever get my butt to fixing up an improvised NAS or similar.

I wrote up a bunch of stuff, but I ended on not knowing what is being said to the point where I may just be repeating moot points.
From what I've seen, NAS acts like just another drive. To me it seems like all you have to do is index that "just another drive" on a separate instance.
Is that not how it works? Is there a point in exporting file lists instead of monitoring the source?
My presumption is no, looking at server options. Never thought about why those are there if you didn't need them for servers.
My only knowledge of NAS behaviour is seeing a few YouTube videos where they happen to show their This PC window and I may be less help and more like spam replying. You tell me.