[OSGeoLive] Disk space and gvSig upgrade

Angelos Tzotsos gcpp.kalxas at gmail.com
Mon Jul 2 08:29:41 PDT 2018


Dear all,

The hard limit for the live system is 4.2GB. The reason is not the iso
size, but the way the live system works: it is based on a FAT file system,
where maximum file size is 4GB. Everything we add is stored on a virtual
file system built on a squashfs file (compressed file system used by ubuntu
for the live system). If we add too much in the customization, the
squashfs goes
over 4GB and cannot be stored in the USB, causing a fatal error.

In order to move away from this limit, we would need to stop distributing a
live system and only produce a VM version.
Another option would be to distribute a light live version in iso format
(excluding big projects from it) and a full version (VM only) that could go
over 4GB in size.

Best,
Angelos


On Mon, Jul 2, 2018 at 5:47 PM, Basques, Bob (CI-StPaul) <
bob.basques at ci.stpaul.mn.us> wrote:

> Jim,
>
> Good points all, and this seems be in line with other list conversations
> related to setting up video tutorials for OSGeo stack items.
>
> bobb
>
>
>
> On Jul 2, 2018, at 9:01 AM, James Klassen <jklassen at sharedgeo.org> wrote:
>
> I can't make the meeting, but I have been wondering for awhile if the
> target should be moved to a 8GB USB stick (or DL DVD).
>
> I am not sure if this holds World wide, but locally 4GB and 8GB USB sticks
> are essentially the same price and 8GB sticks are much easier to find.
> Again locally, from 4GB to 32GB the bigger price difference is USB2 vs USB3
> rather than based on the capacity.
>
> There is certainly benefit in the discipline gained by being forced to be
> efficient with software size.  And I know as someone who has been involved
> in making Live USBs for several conferences that even slight increases in
> cost with limited budgets means less copies can be given out.  But given at
> least my local observations, I no longer see the gain in limiting
> ourselves to 4GB and I worry if we are forced to exclude projects that
> provide good examples of the rich selection of the FLOSS geospatial
> ecosystem based on what may well be an outdated limit.
>
> On Mon, Jul 2, 2018, 07:23 Angelos Tzotsos <gcpp.kalxas at gmail.com> wrote:
>
>> Hi all,
>>
>> We have received a pull request from gvSig to upgrade version from 2.2 to
>> 2.4
>> As can be seen here https://github.com/OSGeo/OSGeoLive/pull/206#
>> issuecomment-401786071 this will double the installation size of gvSig.
>>
>> Lets discuss this tonight, so we can make a decision.
>>
>> Best,
>> Angelos
>>
>> --
>> Angelos Tzotsos, PhD
>> OSGeo Charter Member
>> http://users.ntua.gr/tzotsos
>> _______________________________________________
>> osgeolive mailing list
>> osgeolive at lists.osgeo.org
>> https://lists.osgeo.org/mailman/listinfo/osgeolive
>>
> _______________________________________________
> osgeolive mailing list
> osgeolive at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/osgeolive
>
>
>
>
> "In the end, everything is a gag."- Charlie Chaplin
>
>
>
>
>


-- 
Angelos Tzotsos, PhD
OSGeo Charter Member
http://users.ntua.gr/tzotsos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/osgeolive/attachments/20180702/bebcf315/attachment.html>


More information about the osgeolive mailing list