xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Yang Hongyang <yanghy@cn.fujitsu.com>
To: xen-devel@lists.xen.org
Cc: ian.campbell@citrix.com, wency@cn.fujitsu.com,
	andrew.cooper3@citrix.com, yunhong.jiang@intel.com,
	ian.jackson@eu.citrix.com, eddie.dong@intel.com,
	rshriram@cs.ubc.ca, laijs@cn.fujitsu.com
Subject: [PATCH v18 07/11] xl/remus: cmdline switch to explicitly enable unsafe configurations
Date: Mon, 28 Jul 2014 17:24:01 +0800	[thread overview]
Message-ID: <1406539445-12708-8-git-send-email-yanghy@cn.fujitsu.com> (raw)
In-Reply-To: <1406539445-12708-1-git-send-email-yanghy@cn.fujitsu.com>

By default, network buffering and disk replication are enabled;
checkpoints are replicated to another standby VM.

This patch allows the user to disable any of these features by
explicitly specifying a 'run in unsafe mode' switch when invoking
the 'xl remus' command.  While running Remus in an unsafe mode
makes little sense under normal circumstances, it is useful to be
able to disable one or more features mentioned above for
testing/debugging/profiling purposes.

Unless this option is enabled, it will not be possible to
replicate memory checkpoints to /dev/null (blackhole replication),
disable network buffering or disk replication.

As a starter, the use of blackhole replication now requires that
the unsafe mode be enabled. Subsequent patches will add support
for disabling network buffering and disk replication in a similar
manner.

Signed-off-by: Yang Hongyang <yanghy@cn.fujitsu.com>
Signed-off-by: Shriram Rajagopalan <rshriram@cs.ubc.ca>
---
 docs/man/xl.pod.1           | 15 ++++++++++-----
 tools/libxl/libxl.c         |  5 +++++
 tools/libxl/libxl_types.idl |  1 +
 tools/libxl/xl_cmdimpl.c    | 11 ++++++++++-
 tools/libxl/xl_cmdtable.c   |  7 +++++--
 5 files changed, 31 insertions(+), 8 deletions(-)

diff --git a/docs/man/xl.pod.1 b/docs/man/xl.pod.1
index 30bd4bf..3aedead 100644
--- a/docs/man/xl.pod.1
+++ b/docs/man/xl.pod.1
@@ -441,11 +441,6 @@ B<OPTIONS>
 
 Checkpoint domain memory every MS milliseconds (default 200ms).
 
-=item B<-b>
-
-Replicate memory checkpoints to /dev/null (blackhole).
-Generally useful for debugging.
-
 =item B<-u>
 
 Disable memory checkpoint compression.
@@ -460,6 +455,16 @@ If empty, run <host> instead of ssh <host> xl migrate-receive -r [-e].
 On the new host, do not wait in the background (on <host>) for the death
 of the domain. See the corresponding option of the I<create> subcommand.
 
+=item B<-F>
+
+Run Remus in unsafe mode. Use this option with caution as failover may
+not work as intended.
+
+=item B<-b>
+
+Replicate memory checkpoints to /dev/null (blackhole).
+Generally useful for debugging. Requires enabling unsafe mode.
+
 =back
 
 =item B<pause> I<domain-id>
diff --git a/tools/libxl/libxl.c b/tools/libxl/libxl.c
index 021d77c..6e488ca 100644
--- a/tools/libxl/libxl.c
+++ b/tools/libxl/libxl.c
@@ -797,6 +797,11 @@ int libxl_domain_remus_start(libxl_ctx *ctx, libxl_domain_remus_info *info,
     libxl__domain_suspend_state *dss;
     int rc;
 
+    if (!info->unsafe && info->blackhole) {
+        LOG(ERROR, "Unsafe mode must be enabled to replicate to /dev/null");
+        goto out;
+    }
+
     libxl_domain_type type = libxl__domain_type(gc, domid);
     if (type == LIBXL_DOMAIN_TYPE_INVALID) {
         rc = ERROR_FAIL;
diff --git a/tools/libxl/libxl_types.idl b/tools/libxl/libxl_types.idl
index 25bd8f3..f4cff51 100644
--- a/tools/libxl/libxl_types.idl
+++ b/tools/libxl/libxl_types.idl
@@ -588,6 +588,7 @@ libxl_sched_credit_params = Struct("sched_credit_params", [
 
 libxl_domain_remus_info = Struct("domain_remus_info",[
     ("interval",     integer),
+    ("unsafe",       bool),
     ("blackhole",    bool),
     ("compression",  bool),
     ])
diff --git a/tools/libxl/xl_cmdimpl.c b/tools/libxl/xl_cmdimpl.c
index 01bce2f..3234d45 100644
--- a/tools/libxl/xl_cmdimpl.c
+++ b/tools/libxl/xl_cmdimpl.c
@@ -7175,13 +7175,17 @@ int main_remus(int argc, char **argv)
     memset(&r_info, 0, sizeof(libxl_domain_remus_info));
     /* Defaults */
     r_info.interval = 200;
+    r_info.unsafe = 0;
     r_info.blackhole = 0;
     r_info.compression = 1;
 
-    SWITCH_FOREACH_OPT(opt, "bui:s:e", NULL, "remus", 2) {
+    SWITCH_FOREACH_OPT(opt, "Fbui:s:e", NULL, "remus", 2) {
     case 'i':
         r_info.interval = atoi(optarg);
         break;
+    case 'F':
+        r_info.unsafe = 1;
+        break;
     case 'b':
         r_info.blackhole = 1;
         break;
@@ -7196,6 +7200,11 @@ int main_remus(int argc, char **argv)
         break;
     }
 
+    if (!r_info.unsafe && r_info.blackhole) {
+        perror("Unsafe mode must be enabled to replicate to /dev/null");
+        exit(-1);
+    }
+
     domid = find_domain(argv[optind]);
     host = argv[optind + 1];
 
diff --git a/tools/libxl/xl_cmdtable.c b/tools/libxl/xl_cmdtable.c
index 4279b9f..1e24f1d 100644
--- a/tools/libxl/xl_cmdtable.c
+++ b/tools/libxl/xl_cmdtable.c
@@ -485,13 +485,16 @@ struct cmd_spec cmd_table[] = {
       "Enable Remus HA for domain",
       "[options] <Domain> [<host>]",
       "-i MS                   Checkpoint domain memory every MS milliseconds (def. 200ms).\n"
-      "-b                      Replicate memory checkpoints to /dev/null (blackhole)\n"
       "-u                      Disable memory checkpoint compression.\n"
       "-s <sshcommand>         Use <sshcommand> instead of ssh.  String will be passed\n"
       "                        to sh. If empty, run <host> instead of \n"
       "                        ssh <host> xl migrate-receive -r [-e]\n"
       "-e                      Do not wait in the background (on <host>) for the death\n"
-      "                        of the domain."
+      "                        of the domain.\n"
+      "-F                      Enable unsafe configurations [-b flags]. Use this option\n"
+      "                        with caution as failover may not work as intended.\n"
+      "-b                      Replicate memory checkpoints to /dev/null (blackhole).\n"
+      "                        Works only in unsafe mode."
     },
 #endif
     { "devd",
-- 
1.9.1

  parent reply	other threads:[~2014-07-28  9:24 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-28  9:23 [PATCH v18 00/11] Remus/Libxl: Remus network buffering and drbd disk Yang Hongyang
2014-07-28  9:23 ` [PATCH v18 01/11] libxl: introduce libxl__multidev_prepare_with_aodev Yang Hongyang
2014-07-29 17:21   ` Ian Jackson
2014-07-28  9:23 ` [PATCH v18 02/11] libxl: add support for async. function calls when using libxl__ao_device Yang Hongyang
2014-07-29 17:24   ` Ian Jackson
2014-07-30  8:45     ` Hongyang Yang
2014-07-28  9:23 ` [PATCH v18 03/11] autoconf: add libnl3 dependency for Remus network buffering support Yang Hongyang
2014-07-28  9:23 ` [PATCH v18 04/11] libxl/remus: introduce an abstract Remus device layer Yang Hongyang
2014-08-07 18:30   ` Ian Jackson
2014-08-27  1:46     ` Hongyang Yang
2014-08-27  2:21       ` Ian Jackson
2014-08-27  2:27         ` Hongyang Yang
2014-07-28  9:23 ` [PATCH v18 05/11] libxl/remus: setup and control network output buffering Yang Hongyang
2014-07-28  9:24 ` [PATCH v18 06/11] libxl/remus: setup and control disk replication for DRBD backends Yang Hongyang
2014-07-28  9:24 ` Yang Hongyang [this message]
2014-07-28  9:24 ` [PATCH v18 08/11] xl/remus: cmdline switches and config vars to control network buffering Yang Hongyang
2014-07-28  9:24 ` [PATCH v18 09/11] xl/remus: add a cmdline switch to disable disk replication Yang Hongyang
2014-07-28  9:24 ` [PATCH v18 10/11] libxl/remus: add LIBXL_HAVE_REMUS to indicate Remus support in libxl Yang Hongyang
2014-07-28  9:24 ` [PATCH v18 11/11] MAINTAINERS: update maintained files of Remus Yang Hongyang
2014-08-07  1:17 ` [PATCH v18 00/11] Remus/Libxl: Remus network buffering and drbd disk Hongyang Yang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1406539445-12708-8-git-send-email-yanghy@cn.fujitsu.com \
    --to=yanghy@cn.fujitsu.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=eddie.dong@intel.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=laijs@cn.fujitsu.com \
    --cc=rshriram@cs.ubc.ca \
    --cc=wency@cn.fujitsu.com \
    --cc=xen-devel@lists.xen.org \
    --cc=yunhong.jiang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).