Page 1 of 1

possible small issue in 4.9f?

Posted: Sun Apr 08, 2012 12:54 pm
by ZebraWetter
Hi there,

I just installed a clean v4.9f on a DC01 (I will report later :wink: )
The system information says: v4.9f (build1038) installed.
When meteohub is told to look for a new update, it wants to download v4.9e :shock:
E stands before f in the alphabet, doesn't it? :roll:

Happy easter!
Andre

Re: possible small issue in 4.9f?

Posted: Sun Apr 08, 2012 1:19 pm
by HWTest
I quote from the 4.9f readme:
"Free selectable temperature sensor to use for wind chill and sealevel pressure computation. This feature is still experimental and has potential to produce some new bugs. Cyclic up/downgrade between versions 4.9e and 4.9f allow you to go back if you experience trouble with that."

PS: I envy you for your DC01, I'm still waiting for mine to be delivered :-)

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 12:33 pm
by ZebraWetter
Hm, I didn't test it...
You say, it means, that meteohub recognizes v4.9f and serves now a downgrade to 4.9e and otherwise, if 4.9e is installed, it allows upgrading to 4.9f?
Ok, in this case the "small issue" is not in meteohub, it's in my brain. :roll:

Well, my DC01 has a fresh installation of meteohub since yesterday. Today I want to transfer my meteohub license from my Alix to the DC01. The data and config from the Alix should be moved at the same time. I decided to install a Kingston 16GB SSD, there is enough free capacitiy for the next few years.

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 1:18 pm
by HWTest
That's how I understand it.

I'm planning an ADATA S396S 30GB SSD.
Can you please check, if the partitions are properly aligned?

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 2:52 pm
by admin
I did take care on alligning partitions to full MBs (multiples of 2048 x 512 byte blocks on DC01 platform). But you can check anyway. :D

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 3:09 pm
by HWTest
I think aligning to 4k sectors is sufficient, but aligning to erase block size should be better.
I will check it, when my DC01 finally arrives ...
And if I'm not satisfied, I'll realign it as I did with my SD card in the Sheva :-)

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 3:26 pm
by ZebraWetter
The installed drive is a Kingston SS050S2/16G.

Code: Select all

Disk /dev/sda: 16.0 GB, 16013942784 bytes
255 heads, 63 sectors/track, 1946 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00008000

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1           2       10240   ff  BBT
Partition 1 does not end on cylinder boundary.
/dev/sda2               2         192     1524736   83  Linux
Partition 2 does not end on cylinder boundary.
/dev/sda3             192         217      204800   82  Linux swap / Solaris
Partition 3 does not end on cylinder boundary.
/dev/sda4             217        1947    13897816   83  Linux

Re: possible small issue in 4.9f?

Posted: Mon Apr 09, 2012 4:13 pm
by HWTest
Thank you, it is looking good (4k alignment).

But:

dev/sda2 block size is divisible only by 1024, but the erase block size of most SSDs is 512k, so it is OK
I would shrink /dev/sda4 to 13897728 blocks, so it is divisible by 2048.

Re: possible small issue in 4.9f?

Posted: Tue Apr 10, 2012 3:41 pm
by admin
Partition 4 simply fills up until end of SSD. So I don't think this gives any penalty, regardless if it is on a 512KB border or not.

Re: possible small issue in 4.9f?

Posted: Tue Apr 10, 2012 3:57 pm
by HWTest
I didn't say it's wrong, only what I would do.
I don't think it makes any measurable difference but I would have a better feeling from it :-)