Help with partitions and file systems [SOLVED]

Hi, i have been using Ubuntu for a while, and now i want to move to Arch. I've probed it in a PC and i like so i want to make que change.
But, before installing Arch, I have 2 doubts. I red the beginners guide and also the instalation guide. There it says that if better to have diferents partitions for /, /boot/, /home, /usr, /var, y /tmp
Usually, i alwayes used something like this:
* /boot (32megas)
* /swap (512 megas)
* /root (6 a 8 gigas)
* /home (80 gigas aprox)
It's really better to also have partitions for /var, /usr y /tmp? o some of them? and, in that case, wich size should i give them? because i don't want to make them too small, but i don't want to waste disk space neither.
Adn that takes me to my second question, wich filesystem is better for each partitionn? in many places, i read taht JFS its good for /var or that XFS if better for /home and big files
I thinked to use something like this:
* /boot (ext2)
* / (JFS)
* swap
* /home (XFS)
Is a good design? or should i use other filesystem like reiserfs, etc... and for /var, /usr and /tmp partition, wich one should i use?
Thank you
Ps: This Pc is gonna be a desktop pc
Ps2: sorry for my bad english. it is not my real lenguage
Reason of edit: added the swap partition. I forgot it
Last edited by Thalskarth (2008-12-21 20:11:00)

thanks everybody for the help,
kaola_linux wrote:@Thalskarth - it's better to have /var especially if you're using ext3 or other filesystems which was designed for larger files as your partition for /home and /. Having a seperate partition for /var would be nice (backup purposes and reinstalling without downloading the entire package whole over again). 5gb would be sufficient for your /var, anyway you can always resize it to your needs.
so, a 5gb partition in reiserfs would be OK for a /var
Inxsible wrote:
Lot of people also use XFS which is known to have better performance with huge files. I think EXT3 offers a good balance...because I am never sure whether my home partition will have all huge files or not..same with my external drive...so i just use EXT3
If you have a specific partition for movies or some video/audio editing that you do..you may wanna consider XFS too. I don't do all that...so I have never used XFS. I wouldn't know the exact performance difference between ext3 and XFS.
Yes, in many places i red that XFS is better for big files. But i couldn't fine wich is the meaning of "big file". Does it mean a 200 mg file? or a 4.4gb one??
the same applies to reiserFs, what is a small file? a 1mg one or a 4kb one?
I have alwayes used ext3, i thinked in XFS and JFS just to give them a try.
amranu wrote:I have no idea about better filesystems, all my partitions are ext3 (soon to be ext4)
Inxsible wrote:One thing that makes me wanna keep EXT3 is that EXT4 is coming out (soon?) and you can upgrade from 3 to 4 without having to reformat and having to make backups of your current data.
really is cominf soon? i didn't think in ext4 beacuse many places said that was in development for many years... meybe they were a bit out-of-date
Edit: i search in the wiki and it says that since 11 october 2008, ext4 is "stable" and is been included since kernel 2.6.28 as stable realase, is to early to prove it? or it better to wait a while??
thanks.
and, does anyone try the JFS one?
Last edited by Thalskarth (2008-12-03 00:14:56)

Similar Messages

Need urgent help with service and file removal

Hi,
I need a powershell script that would stop a windows service ,wait for 2 mins, delete a particular file from C:\documents&settings and then start the service again.Is this possible ?
Thanks

Thanks for the reply chen. May be i was not clear in my query
I plan this because a particular file of configmgr is corrupt on workstations. So the service has to be stopped first before the deletion. I need to run this on many computers remotely. The import file will have the machines that need the script to be run.
The script has to run all the four steps on the machines based on the import file and export to a file with the results.How can i modify ?
The sequence will be as follows
Stop-Service -Name BITS -Verbose
Start-Sleep -Seconds 120
Remove-Item -Path C:\documents&settings\***** -Force -Verbose
Start-Service -Name BITS -Verbose
Regards
Sd

Problem with the EFI system partition's file system.

I installed latest yosemite 10.10.2 update - did a verify disk on “WDC WD1001FALS-403AA0 Media” (internal HD on my iMac 21.5-inch, Mid 2011) and got
Problems were found with the partition map which might prevent booting
Error: The partition map needs to be repaired because there’s a problem with the EFI system partition’s file system.
However a verify disk on the volume partition gets a clean bill of health!
I haven't done a 'repair disk' yet. Thought I'd check if anybody else has same message first.
Cheers

I had issues where my MacPro 8-core would not boot after the 10.7.4 upgrade, but my MacBook Air 13" 2011 upgrade worked just fine.
I had to restore my MacPro HD from my TimeMachine back up. Apparently, the Softraid raid drivers that I use on my MacPro are not compatible yet with 10.7.4. See here for more info:
http://macperformanceguide.com/blog/2012/20120511_1-MacOSX-SoftRAID.html
I contacted Softraid and they are already aware of the issues and are working with Apple for a fix.
Cheers.
Bud

MacBookPro SSD Error- ALERT: The partition map needs to be repaired because there's a problem with the EFI system partition's file system.

MacBook Pro OS x Mavericks. I have Sandisk 256GB SSD and I am getting a below error message.
ALERT: The partition map needs to be repaired because there's a problem with the EFI system partition's file system.
Does anyone know how to resolve this error and fix it ? unfortunately I don't have a Time Machine backup. Thanks!

Have you tried to run the verify/repair permissions and verify/repair disk through the recovery disk? If that didn't work then I would reinstall the OS after backing up. If that still doesn't work I would take it in.

Cannot format my iMac. Problem with partition's file

I cannot format my iMac because Disk Utility sends an error: The partition map needs to be repaired because there's a problem with the EFI system partition's file system. Any suggestions?

The problem began when the Time machine excluded my home folder of the backup. I could not include it again in the process (not in system preferences, nor anything else), so I decided to format the HD and make a fresh install of OSX Mavericks. I could not do this because Disk Utility cannot unmount the partition when formatting the HD, and the error arose. I simply cannot fix this, and before giving up and take it to a Mac repair shop, I would like to try by myself.

Using an External Drive with a ReiserFS File System

I hope this is the most appropriate place to ask this question.
I am looking at adding a SimpleShare NAS device (Made by SimpleTech) to my home network. I am told that the hard drive uses the ReiserFS file system (which is a Linux file system). Two questions?
Does anyone have any experience with the ReiserFS file system? Does it play well with OS X?
Anyone familiar with the SimpleShare units? Any feedback good or bad?
Thanks for any suggestions.
David

After doing some research I determined that ReiserFS is a journaled Linux file system and having started to use it, it is working well on the network.

Help with writing and retrieving data from a table field with type "LCHR"

Hi Experts,
I need help with writing and reading data from a database table field which has a type of "LCHR". I have given an example of the original code but don't know what to change it to in order to fix it and still read in the original data that's stored in the LCHR field.
Basically we have two Function modules, one that saves list data to a database table and one that reads in this data. Both Function modules have an identicle table which has an array of fields from type INT4, CHAR, and type P. The INT4 field is the first one.
Incidentally this worked in the 4.7 non-unicode system but is now dumping in the new ECC6 Unicode system.
Thanks in advance,
C
SAVING THE LIST DATA TO DB
DATA: L_WA(800).
LOOP AT T_TAB into L_WA.
ZDBTAB-DATALEN = STRLEN( L_WA ).
MOVE: L_WA to ZDBTAB-RAWDATA.
ZDBTAB-LINENUM = SY-TABIX.
INSERT ZDBTAB.
READING THE DATA FROM DB
DATA: BEGIN OF T_DATA,
             SEQNR type ZDBTAB-LINENUM,
             DATA type ZDBTAB-RAWDATA,
           END OF T_TAB.
Select the data.
SELECT linenum rawdata from ZDBTAB into table T_DATA
     WHERE repid = w_repname
     AND rundate = w_rundate
     ORDER BY linenum.
Populate calling Internal Table.
LOOP AT T-DATA.
APPEND T_DATA to T_TAB.
ENDLOOP.

Hi Anuj,
The unicode flag is active.
When I run our report and then to try and save the list data a dump is happening at the following point
LOOP AT T_TAB into L_WA.
As I say, T_TAB consists of different fields and field types whereas L_WA is CHAR 800. The dump mentions UC_OBJECTS_NOT_CONVERTIBLE
When I try to load a saved list the dump is happening at the following point
APPEND T_DATA-RAWDATA to T_TAB.
T_DATA-RAWDATA is type LCHR and T_TAB consists of different fields and field types.
In both examples the dumps mention UC_OBJECTS_NOT_CONVERTIBLE
Regards
C

I need help with a PDF file that is an image, the "Read Out Loud' option does not work, please help!

I need help with a PDF file that is an image, the "Read Out Loud' option does not work, please help!

You mean an image such as a scanned document?
If that is the case, the file doesn't contain any text for Reader Out Loud to read. In order to fix that, you would need an application such as Adobe Acrobat that does Optical Character Recognition to convert the images to actual text.
You can also try to export the file as a Word document or something else using ExportPDF which I believe offers OCR and is cheaper than Acrobat.

ColdFusion Builder with case sensitive file system on Mac?

Hi there,
I tried to install CF Builder on my MacBook Pro (Intel), OS-X 10.5.7 with case sensitive file system. Installation went smoothly, but, when I try to launch CF Builder, I get an error saying that the "architecture is not supported". Is it just nor possible to have CF Builder on a case sensitive fs?
Thank you
Chris

I am having the same issue and logged in here to see what the deal is.
After I installed cfbuilder my flex builder crapped out as well.
The cfbuilder logs (which were conveniently scattered throughout my system for no reason) report that I am running:
java.vm.version == 1.5.0_19-137
But if I run "java -version" in terminal I get:
daves-macbook-pro-2:~ Dave$ java -version
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03-211)
Java HotSpot(TM) 64-Bit Server VM (build 11.3-b02-83, mixed mode)
"Is JDK 1.5 available in the system"
Definale "available"...
Does it exist? yes
Is it set to default? no

How many partitions and operating systems can you line up in BootCamp?

A quick question, hopefully with a quick answer.
How many partitions and operating systems can you line up in BootCamp?
Are we limited to just 2, or can we have more?
Say, for example, Mac, Win XP, Win Vista and Linux maybe?
Thanks in advance.

VPC does not run on Intel Macs. If you wish to use virtualization on an Intel Mac see this:
Windows on Intel Macs
There are presently several alternatives for running Windows on Intel Macs.
1. Install the Apple Boot Camp software. Purchase Windows XP w/Service Pak 2 or Vista. Follow instructions in the Boot Camp documentation on installation of Boot Camp, creating Driver CD, and installing Windows. Boot Camp enables you to boot the computer into OS X or Windows.
2. Parallels Desktop for Mac and Windows XP, Vista Business, or Vista Ultimate. Parallels is software virtualization that enables running Windows concurrently with OS X.
3. VM Fusionand Windows XP, Vista Business, or Vista Ultimate. VM Fusion is software virtualization that enables running Windows concurrently with OS X.
4. CrossOver which enables running many Windows applications without having to install Windows. The Windows applications can run concurrently with OS X.
5. VirtualBox is a new Open Source freeware virtual machine such as VM Fusion and Parallels that was developed by Solaris. It is not yet fully developed for the Mac - some features are not yet implemented - but it does work otherwise.
6. Last is Q. Q is a freeware emulator that is compatible with Intel Macs. It is much slower than the virtualization software, Parallels and VM Fusion.
Note that Parallels and VM Fusion can also run other operating systems such as Linux, Unix, OS/2, Solaris, etc. There are performance differences between dual-boot systems and virtualization. The latter tend to be a little slower (not much) and do not provide the video performance of the dual-boot system.
See MacTech.com's Virtualization Benchmarking for comparisons of Boot Camp, Parallels, and VM Fusion.
Boot Camp is only available with Leopard. The Boot Camp Beta that was used with Tiger has expired and is no longer available for use. So contrary to the other poster's comment, Boot Camp isn't truly "free." You must purchase Leopard to get it.
You can upgrade XP to Vista without having to make any changes to the Boot Camp partition. Vista can either perform an upgrade install or erase the Windows volume before installing Vista. You don't need to do anything to change the partition via Boot Camp. However, if you need to enlarge the partition, then the only way that can be done is to first use Boot Camp to remove the partition. Then use Boot Camp to make a new, larger partition. All data from the Windows partition will be lost.

MOVED: [Athlon64] Need Help with X64 and Promise 20378

This topic has been moved to Operating Systems.
[Athlon64] Need Help with X64 and Promise 20378

I'm moving this the the Administration Forum. It seems more apporpiate there.

How to bulk import data into CQ5 from MySQL and file system

Is there an easy way to bulk import data into CQ5 from MySQL and file system? Some of the files are ~50MB each (instrument files). There are a total of ~1,500 records spread over about 5 tables.
Thanks

What problem are you having writing it to a file?
You can't use FORALL to write the data out to a file, you can only loop through the entries in the collection 1 by 1 and write them out to the file like that.
FORALL can only be used for SQL statements.

Help with count and sum query

Hi I am using oracle 10g. Trying to aggregate duplicate count records. I have so far:
SELECT 'ST' LEDGER,
CASE
WHEN c.Category = 'E' THEN 'Headcount Exempt'
ELSE 'Headcount Non-Exempt'
END
ACCOUNTS,
CASE WHEN a.COMPANY = 'ZEE' THEN 'OH' ELSE 'NA' END MARKET,
'MARCH_12' AS PERIOD,
COUNT (a.empl_id) head_count
FROM essbase.employee_pubinfo a
LEFT OUTER JOIN MMS_DIST_COPY b
ON a.cost_ctr = TRIM (b.bu)
INNER JOIN MMS_GL_PAY_GROUPS c
ON a.pay_group = c.group_code
WHERE a.employee_status IN ('A', 'L', 'P', 'S')
AND FISCAL_YEAR = '2012'
AND FISCAL_MONTH = 'MARCH'
GROUP BY a.company,
b.district,
a.cost_ctr,
c.category,
a.fiscal_month,
a.fiscal_year;
which gives me same rows with different head_counts. I am trying to combine the same rows as a total (one record). Do I use a subquery?

Hi,
Whenever you have a problem, please post a little sample data (CREATE TABLE and INSERT statements, relevant columns only) from all tables involved.
Also post the results you want from that data, and an explanation of how you get those results from that data, with specific examples.
user610131 wrote:
... which gives me same rows with different head_counts.If they have different head_counts, then the rows are not the same.
I am trying to combine the same rows as a total (one record). Do I use a subquery?Maybe. It's more likely that you need a different GROUP BY clause, since the GROUP BY clause determines how many rows of output there will be. I'll be able to say more after you post the sample data, results, and explanation.
You may want both a sub-query and a different GROUP BY clause. For example:
WITH    got_group_by_columns     AS
     SELECT a.empl_id
     ,     CASE
                    WHEN c.category = 'E'
              THEN 'Headcount Exempt'
                    ELSE 'Headcount Non-Exempt'
            END          AS accounts
     ,      CASE
                    WHEN a.company = 'ZEE'
                    THEN 'OH'
                    ELSE 'NA'
            END          AS market
     FROM              essbase.employee_pubinfo a
     LEFT OUTER JOIN mms_dist_copy             b ON   a.cost_ctr     = TRIM (b.bu)
     INNER JOIN       mms_gl_pay_groups        c ON   a.pay_group      = c.group_code
     WHERE     a.employee_status     IN ('A', 'L', 'P', 'S')
     AND        fiscal_year           = '2012'
     AND        fiscal_month          = 'MARCH'
SELECT    'ST'               AS ledger
,       accounts
,       market
,       'MARCH_12'          AS period
,       COUNT (empl_id)      AS head_count
FROM       got_group_by_columns
GROUP BY accounts
,            market
;But that's just a wild guess.
You said you wanted "Help with count and sum". I see the COUNT, but what do you want with SUM? No doubt this will be clearer after you post the sample data and results.
Edited by: Frank Kulash on Apr 4, 2012 5:31 PM

I have interface M Audio firewire 400 new Mac Pro has firewire 800 Do you think an adaptor will provide the same quality? Will the M audio 400 work with lion and new system? Thanks so much

I have interface M Audio 400 with 400 firewire new Mac Pro has firewire 800 Do you think an adaptor will provide the same quality? Will the M audio 400 work with lion and new system or do I need new software? Thanks so much

If you have a little Firewire adaptor, everything should work fine. It does on my system.

Partition scheme and file systems in dual boot environment [Solved]

This machine is a general purpose desktop computer used by a single user. The plan is to run Arch primarily and have Ubuntu for backup. I have some experience with Arch and various other linux distros and usually just use a single partition for / with a separate /home partition. On my primary machine, however, I've accumulated a few disks and am wondering how I could best make use of what I have. I may eventually add a 256gb SSD or replace the 80gb drive with it, but currently this is what I have:
8gb SSD
80gb SSD
80gb HDD
1.5tb HDD
18gb RAM
Previously I kept almost everything on the 80gb SSD, including the home directory with individual folders for large files (VM's, music, video) stored on the 1.5tb drive and symlinked into my home directory. This worked very well, except the my home directory still wanted to grow fairly large over time as things accumulate (one culprit was my bitcoin wallet). If I did this again, it would be Arch on the 80gb SSD (probably using BTRFS), Ubuntu on the 80gb HDD (probably EXT4 or BTRFS), and a few different partitions on the 1.5tb drive, mostly BTRFS, depending on what will be stored.
My other thought was to install both operating systems to the 80gb SSD for maximum speed while putting the home directories on the 80gb HDD or on separate partitions of the 1.5tb drive.
In either case I'll probably make a /scratch mounted as a tmpfs to make better use of RAM.
So questions:
How much does speed matter generally for a home partition?
Is BTRFS a good choice for this kind of system? More trouble than it's worth?
Would an LVM make sense in this case? I'd assume there would be performance issues if I tried mixing SSD's and HDD's in an LVM.
What should I use the small 8gb SSD for?
Sorry if I've mixed terminology some here, I'm still learning!
Last edited by spurious_access (2013-07-12 15:51:37)

1) For most user data, speed is not really of consequence. But what can matter is the speed of all the configuration files located in your home folder. They are small, so decreased seek times are going to have a positive effect here. This is the same reasoning as to why you would want the system configuration files in /etc to be on the SSD (not to mention you'd have a hell of a time separating /etc).
2) Btrfs is f*cking awesome!
3) LVM could make sense, but it depends on what you want specifically. The only reason you might have a performance hit with LVM2 is if you tried to apply some kind of RAID to it. That is, striping and mirroring in LVM2 terms. Obviously, vastly different speeds are going to do limited by whatever is slowest. But the nice thing about LVM2 is that you can tell lvcreate to ensure that your logical volume is to be created all on one disk. But btrfs is actually designed to handle disks of different sizes and to some extent, different speeds.
4) If the 8GB SSD is fast, then maybe you could use it as a nice root filesystem. I honestly don't think I have ever had my root get any larger than 5GB. But then I don't run a full DE either, but I also don't keep my system as mega-slim as possible like Trilby. It might make sense to LVM over the two SSDs and maybe have a separate volume group over the two HDDs. Or have one btrfs filesystem spanning the two SSDs and one spanning the two HDDs.
Ultimately, the needs of your personal usage style is going to dictate what the actual best configuration of your machine is going to be. But this hodge-podge of disks is defintiely not a terrible thing, and can actually probably be used quite efficiently. It is totally up to you.
As a side note (in regard to my "btrfs is f*cking awesome" comment above), I tried btrfs about a year ago. It was nice, but it was not quite where I had hoped. But after doing something stoopid on my machine, and having to redo part of my setup, I decided to just rsync my system to another disk, create a btrfs pool, then rsync it back. I couldn't be happier with my decision. It is truly amazing the rate at which btrfs is progressing. New features, more stability, faster speeds, it is all coming along quite nicely.

Help with partitions and file systems [SOLVED]

Similar Messages

Maybe you are looking for