From: Iustin Pop Date: Fri, 4 Feb 2011 09:25:03 +0000 (+0100) Subject: Update NEWS file for the 0.3.0 release X-Git-Tag: htools-v0.3.0^0 X-Git-Url: https://code.grnet.gr/git/ganeti-local/commitdiff_plain/a7c0793556eeae5ce04f33d7eeab66abb1604f42?ds=sidebyside Update NEWS file for the 0.3.0 release Signed-off-by: Iustin Pop Reviewed-by: Michael Hanselmann --- diff --git a/NEWS b/NEWS index 1783ea1..b92d0c6 100644 --- a/NEWS +++ b/NEWS @@ -2,6 +2,45 @@ Ganeti-htools release notes =========================== +Version 0.3.0 (Fri, 04 Feb 2011) +-------------------------------- + +A significant release that breaks compatibility with Ganeti versions +below 2.4 due to the node group changes. Only the RAPI backend can talk +to older clusters, but it is recommended to use this version only with +Ganeti 2.4. + +All commands are now multi-group aware (but to various degrees), so +allocation, balancing and capacity calculation respects the group layout +and will not create “broken” instances by using nodes from different +groups. + +For a regular, single-group cluster, no changes should be directly +visible to the users. A multi-group cluster however will change some +things slightly: + +- hbal will require a target group to operate on (no cluster-wide + balancing yet) +- evacuation of (DRBD) instances from a node will be restricted to nodes + in the same group, as inter-group moves are not implemented yet +- capacity, while showing correct data, will not give per-group details + yet + +There are other changes in this release: + +- fixed a long-standing bug in hscan related to node memory data +- changed the text backend format, which unfortunately invalidates old + files +- error handling improvements, so that invalid input data reports better + where the error is +- the simulation backend changes its syntax, now it takes the allocation + policy too, and can generate multiple groups +- (internal) man page generation moved to pandoc from hand-written, + which is helpful as it can also generate HTML versions +- the balancing algorithm has been changed to work in parallel, if the + code is linked against the multi-threaded runtime; this gives a very + good speedup (~80% on 4 cores, ~60-70% of 12 cores) + Version 0.2.8 (Thu, 23 Dec 2010) --------------------------------