http://www.openegg.org/forums/forums/show/4.page JForum - http://www.jforum.net http://www.openegg.org/forums/posts/preList/976/9353.page http://www.openegg.org/forums/posts/preList/976/9353.page GMT Still using flexraid and proud

I came back here to give some feedback hoping to make flexraid eve better
I'll be posting it on mantis where it belongs too

Just a small case that just occurred to me to explain the request

'cause of some performance issues I was forced to exchange one of my storage server HD's with one in other machine I have
no problem with it, but on the mounting back, I just messed up and let the PPU HD cable loose
server restared fine but I didn't noticed the issue 'cause the PPU is exclusive for this purpose
than I decided to do a quick validade throught the webUI and got a fail due to the missing unit
just then I figured the problem out
reopened the server, put the cable back and got it running fine
with the server running ok with all the drives detected I went back to do the quick validate and got a fail 'cause a validade task was running

the question here is: there is no clear, explicit and easy feedback on whatever flexraid is doing

some small system tray app would be great for this
maybe some color based icon

red for some fail (validation or any other important fail), maybe with a popping toast...
yellow for some actitity (scheduled or demanded action as verify, update or so)
and green if flexraid is idle and everything is fine since the last verify (quick or standard)

right clicking the tray icon would bring a context menu with some controls like stop or pause a running task, or to manually start one if idle
and a option to open the full interface with the webUI on localhost
maybe a link to the log page on webIU some other forum friend asked for

such small app would be very nice and important to users (newbies or not) to have live and easy feedback and control of flexraid

as I said, hope it helps to make flexraid even better
keep up the good job guys

[]s]]> http://www.openegg.org/forums/posts/preList/957/9162.page http://www.openegg.org/forums/posts/preList/957/9162.page GMT
But would it be possible to do this

Set up an array of data with redundancy equivalent to raid 5 so you have the live data as protected as possible

But then treat the entire raid5 array as though its a virtual physical disk and have an actual physical disk set up in an intelligent raid1 style configuration

This would in effect be a back up of the raid5 array. But would (I think) have the following advantages and could/should/would work as follows



Once the relationship was created the extra drive lets assume its a USB external would be populated with a mirror of the raid array initially. But it could then be removed.

This mirror drive would be instantly recognised for what it is when connected to the system, but wouldnt cause a problem when removed

The system would then build a log of all changes since the last equalisation between the array and the USB Raid 1 mirror, so the next time it was plugged in only changes would need to be synchronised

The external drive could function as a raid1 mirror for several raid5 active arrays across multiple machines with each partition/folder recognised by the system/raid5 array its paired with when inserted

If a raid5 array failed to the extent of data being lost then once storage space was replaced and added to the raid5 array the removeable mirror drive could then be reatteched and used to repopulate the data in the raid5, or even to create identical raid5 arrays on several machines, or a new, larger one in a machine if all the disks were updated

So in effect you would have perhaps 4x500gb folders across 4 drives under flexraid on a machine mounted as one virtual drive lets say Drive M:

You plug in a USB/Esata 2tb external drive which is its removeable mirror, flexraid detects its been inserted and checks the changelog for all files altered, added or deleted since the last insertion and adds the new ones, adds the changed ones but perhaps keeps the previous version too?? And also removes deleted ones but doesnt delete them so that changed and deleted files can be restored to the previous version if wanted (space dependant obviously) all done in the background

The extra drive can then either be left en situ for added security whilst working on files, or removed when the psuedo raid1 mirroring is complete. And I guess for added flexbility the syncrhonisation wouldnt necessarily need to be done all at once and could either be done as a background task or even in portions allowing the removal and reattaching of the mirror drive at which point flexraid would see how far it got through synchronising and would carry on from where it left off. Any newly added changes would just be at the end of the log anyway so it wouldnt really care when a change occured, only that it hasnt updated the removeable mirror with that change yet if that makes sense?


I know a fairly similar thing can just be done with back-up software. But from what I have read about flexraid it would be flexible enough to layer raid types and handle the topic of back ups far more seamlessly and without the need for additional software by doing something along these lines

I probably havent really explained this very well so apologies if my idea is hard for the more technically minded folksies to understand, and also if its either a dumb idea or something that already exists

This would seem to fit in with the flexraid ethos of protecting data rather than datablocks. And would do away with the need for additional back up software whilst seeming to be an easy way to repopulate a totally dead raid5 type array if all or most data was lost due to more failures than the raid type could allow

Just a thought, and hopefully not THAT stupid an idea lol, but it just seemed like a good idea when it popped into my noggin

Once again apologies if I havent explained it very clearly

Mike]]> http://www.openegg.org/forums/posts/preList/940/9018.page http://www.openegg.org/forums/posts/preList/940/9018.page GMT
But in general, topics in the wiki seem organized out of order, and are often answered better some place else. Maybe some of the users would be interested in house keeping. Eg. I'd be willing to put in 2-3 hours, though I don't know how much I could really contribute.]]> http://www.openegg.org/forums/posts/preList/906/8826.page http://www.openegg.org/forums/posts/preList/906/8826.page GMT http://www.openegg.org/forums/posts/preList/897/8782.page http://www.openegg.org/forums/posts/preList/897/8782.page GMT
It would be nice to have progress bar that reports the current completion percentage of update / verify tasks (any task that can last 10s of minutes or hours).
That could be really helpful...

Btw, since mantis does not support feature requests, only bugs, here is still the place for feature requests?
]]> http://www.openegg.org/forums/posts/preList/891/8706.page http://www.openegg.org/forums/posts/preList/891/8706.page GMT User will select available drives letters that they want to sample. Each will require about 10GB of empty space. (maybe the size is an option, eg: 1/5/10GB)
User will select 1 of the total selected drives as a parity drive.
Then FR will generate 10GB or random for the 1st drive, log the timing.
After it finish, then read the data off the 1st drive read it to null and time it. So now we have read/write speed for 1st drive.
repeat it to all the selected drive(including the parity drive).
At this point, all the drives now would have a sample of read/write speed to show.
Then delete the random data written on the parity drive and proceed to benchmarking RAID option.
The system then go through all the mention, em, a bit of small algorithm, and adjust one option at a time, each creating parity and log the timing.
The system pick a few of the best timing available to show user and present with the option.(or maybe a feature to save those options as prefer option to use later?)
Upon completion of this benchmark mode, all the files created during this process get delete.

I mean, all the function is already there. This should be relatively easy as there is nothing new being invented. TrueCrypt has benchmark also, doing similar thing, so why not? This way, not only user will get the best performance they can get, but also reduce the likelihood of users making poor configuration choices on a large data later on, which not only lead to poor performance but also unnecessary questions on the forum. Perhaps if the data(log) is being collected in a useful manner
, maybe user can volunteer to upload the said log/stat file that can help other comparing hardware and performance. Something like raid creation timing, harddrive model(but no serial number) read/write time, CPU & controllers make/models, memory size and maybe controller drivers version; those should be in the log as well. I think those are useful information.]]> http://www.openegg.org/forums/posts/preList/845/8332.page http://www.openegg.org/forums/posts/preList/845/8332.page GMT
i recently start to reuse flexraid again and would have some ideas/wishes

feature request #1:
i am using hotswap sata disks and truecrypt on my system which leads to the problem that its kinda tricky to ensure that a certain hdd/partition always gets the same drive letter when mounted in windows.. this leads of course to problems then when those disks are used in a dru/ppu by flexraid.

thats why i would like to propose/request the following:
after the initial setup flexraid should be able to identify disks and associate them to the intended dru/ppu automatically once flexraid runs.

to clear up:
initial setup
d:\ hdd1 -> DRU1
e:\ hdd2 -> DRU2
f:\ hdd3 -> PPU1

later flexraid is also able to notice when the disks are found like
d:\ hdd3 -> PPU1
e:\ hdd2 -> DRU2
f:\ hdd1 -> DRU3

the identification is maybe tricky .. im not sure if there is a unique id you can read out from an disk/partition for this. on the other hand you could also create a small file in the root of the used disks which contains the needed data .. like once flexraid starts it scans the root of every drive for this file..
(problem then maybe if only a subfolder is used for an dru/ppu and not the whole disk)

this would be nice for the view aswell of course


feature request #2:
another feature wish would be the following:
instead of the complex setup (especially for many disks) one could also imagine a simplified setup like
1. tell flexraid the disks to use (lets say 12x 2TB)
2. set level of parity protection (lets say 4 TB)
3. set level of spanning (lets say option 1 = none | option 2 = 2 disks)
now flexraid has all it needs to setup "automatically" :
option1 -> 10 DRU's with 2TB each, 2 PPU's with 2TB each
option2 -> 5 DRU's with 4TB each, 1 PPU with 4TB (each PPU/DRU = 2disks)
the setup is easy for flexraid i guess.. especially when there are enough empty disks for the ppus

the example is simple.. but flexraid could setup automatically even for complex situations .. like for lots of diff sized disks.
actually the whole concept of having dru/ppu could be hidden from the user i think, it wouldnt only simplify the usage but probably also be less confusing for people who dont want/can/whatever get into the topic


sorry .. this became longer than intended haha ... hope it wasnt too much to get through and understandable still.

greets and thanks for a nice product already!
mikk]]> http://www.openegg.org/forums/posts/preList/842/8325.page http://www.openegg.org/forums/posts/preList/842/8325.page GMT
It would be nice if there was a button that would collect data into a file.

The data would be.

Version and build number of FlexRaid.
Os version, service pack number.
Name of the system.
Number and type of cores.
Number of disks.
FlexRaid log files
FlexRaid config files
FlexRaid core dump (if any)

I'll be running this software in a number of machine, real and virtuial and this type of a feature would really help.

]]> http://www.openegg.org/forums/posts/preList/839/8296.page http://www.openegg.org/forums/posts/preList/839/8296.page GMT
It would then be useful to be able to sort and filter on date and message type (e.g. error, warning, info..).

Config of the log file behavior should then also belong here. Probably other more experienced users have useful input to this also.]]> http://www.openegg.org/forums/posts/preList/807/8140.page http://www.openegg.org/forums/posts/preList/807/8140.page GMT I've been using Flexraid for about 6 months now and, specially the 2.0 version is AWESOME!!
The project lacks a lot of documentation and some old info may be misleading, but I this topic is not about it

One of the major causes for me to be using flexraid is the storage pool. Never used WHS before, but it really helps when it comes to manage my media library
The question is, storage pool is kinda a nightmare to configure. And even when configured, some times it just don't do what it's told to
Since my configuration is quite simple (multiple DRUs with the same folder structure), and as I don't wanna any data split among them, after reading a little I've found the Auto-Minimized-Folder-Split-Priority option that did the job very simply.

The question here is:
Flexraid WebUI, in the raid configuration, has a nice drag and drop tree explorer interface. This exactly same interface can make the storage pool configuration LOTS easier.
The idea is, as in the raid config, the tree explorer in the left, in the right, just another one, but built on-the-fly.
A combo on top to enter the device letter, some icons to add, remove and rename (manage) virtual and real folders folders.
The config would be as easy as draggin the folders from the left to the right tree
Maybe some radio buttons to choose between Auto-Minimized-Folder-Split-Priority and the other new types of merge.

The tecnology is already packed in, so I guess implementing it should not be that hard.
Hope this helps to make flexraid even better

And please, as a developer myself, I HATE documenting stuff. But costumers and new users depend on that to be able to use any system
So, please, don't let the documentation aside. It could REALLY impact on the number of users of flexraid just as much as I hope version 2.0 did (MUCH easier to go)

Keep up the good job

[]s]]> http://www.openegg.org/forums/posts/preList/805/8118.page http://www.openegg.org/forums/posts/preList/805/8118.page GMT
I tested FlexRAID 2.0 preview8 to see what a restore did with regards to date/time preservation.
I saw that they were preserved for files but not for directories. I really wish for them to be preserved
for both files and directories. The test was made on Windows.

The new GUI was really nice compared to since last I tested it. Keep up the good work. Thankyou!

Best regards

/Thomas]]> http://www.openegg.org/forums/posts/preList/773/7875.page http://www.openegg.org/forums/posts/preList/773/7875.page GMT
Basically like this
[code]-X:\
|-<Music>
|-#Pictures#

-X:\Music
|-*C:\Drives\F
|-*C:\Drives\G

-X:\Pictures
|-*C:\Drives\H
|-*C:\Drives\I
|-*C:\Drives\J
[/code]Basically if I were to Put a file in Pictures It would copy the Files to Two of the Three Locations, and if those run out of space it will pick two locations from the pool...

Or you could implement it like this[code]-X:\
|-<Music>
|-<Pictures>

-X:\Music
|-*C:\Drives\F
|-*C:\Drives\G

-X:\Pictures
|-#C:\Drives\H
|-*C:\Drives\I
|-*C:\Drives\J
[/code]Where If i put a file in Pictures it ALWAYS puts the file in H and then puts another in I or J

]]> http://www.openegg.org/forums/posts/preList/680/7121.page http://www.openegg.org/forums/posts/preList/680/7121.page GMT
For example if I have my videos stored in:
D:\Videos
E:\Videos

Instead of loading up the disk listed first as the View\Videos folder, why not have it balance the storage between the 2 disks. This does not have to balance the storage after it has been written like WHS DE, but I am thinking about looking at which folder has the smallest amount of used space and write to there.

That way it will help balance out the wear and tear on the disks and allow for higher I/O for items stored on separate spindles.]]> http://www.openegg.org/forums/posts/preList/629/6645.page http://www.openegg.org/forums/posts/preList/629/6645.page GMT This feature would be similar to what XBMC does if you select multiple directories as source.
Read-only because if you merge directories like this it's hard to know where to write files.
]]> http://www.openegg.org/forums/posts/preList/593/6299.page http://www.openegg.org/forums/posts/preList/593/6299.page GMT
I remember reading a few things about caching the view in a few threads here and there and wondered what the status of this was now - so thought we could track it in a thread here.

I think it is a particularly good idea due to:
1. In large drive setups it could stop drives from spinning up via a client using windows explorer to list files that a sleeping drive has on it
2. Well I think (1) is mainly the reason :) Faster response to clients could be a small 2nd benefit?

In terms of the issue around people writing to a drive outside the view:
1. Have an option in the viewconfig.txt where we can enable/disable the cache based on our position on this (I know in my setup I will only ever write to the view or to a share on the view)]]> http://www.openegg.org/forums/posts/preList/513/5340.page http://www.openegg.org/forums/posts/preList/513/5340.page GMT
]]> http://www.openegg.org/forums/posts/preList/493/5141.page http://www.openegg.org/forums/posts/preList/493/5141.page GMT
- Let's get crazy: Implement tab completion and...

- ...partial command parsing (i.e. parse the command correctly when there is a single match of the typed characters, like Cisco for example).

- Since most people try "exit" first before remembering dot, why not just add "exit" to the command list if it's not too hard.

]]> http://www.openegg.org/forums/posts/preList/473/4914.page http://www.openegg.org/forums/posts/preList/473/4914.page GMT
anyone like the idea ?]]> http://www.openegg.org/forums/posts/preList/466/4865.page http://www.openegg.org/forums/posts/preList/466/4865.page GMT http://www.openegg.org/forums/posts/preList/440/4676.page http://www.openegg.org/forums/posts/preList/440/4676.page GMT http://www.openegg.org/forums/posts/preList/424/4463.page http://www.openegg.org/forums/posts/preList/424/4463.page GMT
From Wikipedia: http://en.wikipedia.org/wiki/Filesystem_in_Userspace
[quote]FUSE is particularly useful for writing virtual file systems. Unlike traditional file systems that essentially save data to and retrieve data from disk, virtual file systems do not actually store data themselves. They act as a view or translation of an existing file system or storage device.[/quote]

]]> http://www.openegg.org/forums/posts/preList/403/4169.page http://www.openegg.org/forums/posts/preList/403/4169.page GMT http://www.openegg.org/forums/posts/list/400.page[/list]



]]> http://www.openegg.org/forums/posts/preList/401/4151.page http://www.openegg.org/forums/posts/preList/401/4151.page GMT Thanks]]> http://www.openegg.org/forums/posts/preList/400/4150.page http://www.openegg.org/forums/posts/preList/400/4150.page GMT http://www.openegg.org/forums/posts/preList/385/3959.page http://www.openegg.org/forums/posts/preList/385/3959.page GMT http://www.openegg.org/forums/posts/preList/285/2980.page http://www.openegg.org/forums/posts/preList/285/2980.page GMT
I understand that some people use hidden files as an exclude, but would it be possible to have an option set in the config file to include all hidden files, while leaving them excluded by default?

With Flexraid as it stands I am not able to backup over 1000 movies worth of metadata and DVD covers, because my metadata manager sets then to hidden by default.]]> http://www.openegg.org/forums/posts/preList/239/2355.page http://www.openegg.org/forums/posts/preList/239/2355.page GMT http://www.vilett.com/disParity/forum/topic.php?id=57 , although not for flex, this does show an attempt to create a gui for similar purposes. I thought it was worth showing.

Ben Ogilvie]]> http://www.openegg.org/forums/posts/preList/226/2190.page http://www.openegg.org/forums/posts/preList/226/2190.page GMT
Quick-validate, validae and verify, all just check if your EXISTING data are ok.
There is nothing to check if rsync is needed.

You just have to "know" which is ok if you update the files yourself only, but not ok if other people also update the data, or if scheduled recordings take place etc.

(BTW Brahim you have missed two of my PM, hope you are ok)

]]> http://www.openegg.org/forums/posts/preList/198/1911.page http://www.openegg.org/forums/posts/preList/198/1911.page GMT I trust those metadata also hold the pool details (even if the pools are more than one).
If not, then I guess they are stored somewhere (like some system folder or FlexRAID install folder or you tell us where). I don't know.

This means that to be able to recover from a problem, you need the parity AND the metadata AND the configs (if the metadata do not carry the config details). What happens if the problem is exactly where the metadata and/or config files are? How does FlexRAID kickstart then?

Currently the user has to have a backup somewhere. A RAID system that asks you for a backup? Hm...

How should it be?

1) Default the metadata (and config files if needed) on a different folder from the parity (preferably in some system folder accessible by all users).

2) "Mirror" the metadata (and config files if needed) IN THE PARITY file(s).

What do you gain from that?

If you loose the parity folder (or disk if we talk about full disks), then the metadata (or config) needed to rebuild the array and it's parity, are somewhere else.

If you loose the disk (or folder) where the metadata (or config) are normally stored, then the parity file(s) can recover that too (also check my PM to see some extension to this).

In other words, you have two points of failure not only for the data, but also for the arrays themselves!

That's how it should be.

Hardware RAID solutions (talking about expensive server solutions not 50 euro home solutions), use (1) local flash memory and (2) the array disks themselves, to store the array configuration.
So whichever disk you need to replace, doesn't mean you lose your configuration (of course). On the other hand if you need to replace your RAID card (with a same model of course), you STILL don't lose your configuration because the new card looks in the disks connected on it to see if a configuration already exists. Finds it and uses it.

This is what I am trying to emulate with my idea.

What do you think?

(and easy to implement even for 1.3 final)

]]> http://www.openegg.org/forums/posts/preList/193/1874.page http://www.openegg.org/forums/posts/preList/193/1874.page GMT