diff mbox series

[v5] docs/designs: re-work the xenstore migration document...

Message ID 20200428150624.265-1-paul@xen.org (mailing list archive)
State New, archived
Headers show
Series [v5] docs/designs: re-work the xenstore migration document... | expand

Commit Message

Paul Durrant April 28, 2020, 3:06 p.m. UTC
From: Paul Durrant <pdurrant@amazon.com>

... to specify a separate migration stream that will also be suitable for
live update.

The original scope of the document was to support non-cooperative migration
of guests [1] but, since then, live update of xenstored has been brought into
scope. Thus it makes more sense to define a separate image format for
serializing xenstore state that is suitable for both purposes.

The document has been limited to specifying a new image format. The mechanism
for acquiring the image for live update or migration is not covered as that
is more appropriately dealt with by a patch to docs/misc/xenstore.txt. It is
also expected that, when the first implementation of live update or migration
making use of this specification is committed, that the document is moved from
docs/designs into docs/specs.

NOTE: It will only be necessary to save and restore state for active xenstore
      connections, but the documentation for 'RESUME' in xenstore.txt implies
      otherwise. That command is unused so this patch deletes it from the
      specification.

[1] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/designs/non-cooperative-migration.md

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
---
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: George Dunlap <george.dunlap@citrix.com>
Cc: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Jan Beulich <jbeulich@suse.com>
Cc: Julien Grall <julien@xen.org>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Wei Liu <wl@xen.org>

v5:
 - Addressed comments from Julien
 - Dropped 'mfn' from shared ring 'conn-spec' as its feasibility/utility is
   debatable

v4:
 - Move path-len and value-len earlier in NODE_DATA

v3:
 - Address missed comments from Juergen

v2:
 - Address comments from Juergen
---
 docs/designs/xenstore-migration.md | 469 +++++++++++++++++++----------
 docs/misc/xenstore.txt             |  17 --
 2 files changed, 307 insertions(+), 179 deletions(-)

Comments

Julien Grall May 5, 2020, 1:01 p.m. UTC | #1
Hi Paul,

On 28/04/2020 16:06, Paul Durrant wrote:
> From: Paul Durrant <pdurrant@amazon.com>
> 
> ... to specify a separate migration stream that will also be suitable for
> live update.
> 
> The original scope of the document was to support non-cooperative migration
> of guests [1] but, since then, live update of xenstored has been brought into
> scope. Thus it makes more sense to define a separate image format for
> serializing xenstore state that is suitable for both purposes.
> 
> The document has been limited to specifying a new image format. The mechanism
> for acquiring the image for live update or migration is not covered as that
> is more appropriately dealt with by a patch to docs/misc/xenstore.txt. It is
> also expected that, when the first implementation of live update or migration
> making use of this specification is committed, that the document is moved from
> docs/designs into docs/specs.
> 
> NOTE: It will only be necessary to save and restore state for active xenstore
>        connections, but the documentation for 'RESUME' in xenstore.txt implies
>        otherwise. That command is unused so this patch deletes it from the
>        specification.
> 
> [1] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/designs/non-cooperative-migration.md
> 
> Signed-off-by: Paul Durrant <pdurrant@amazon.com>
> Reviewed-by: Juergen Gross <jgross@suse.com>

Acked-by: Julien Grall <jgrall@amazon.com>

I will wait until tomorrow afternoon (BST) to give an opportunity to 
other to comment.

Cheers,
Jürgen Groß May 5, 2020, 1:13 p.m. UTC | #2
On 05.05.20 15:01, Julien Grall wrote:
> Hi Paul,
> 
> On 28/04/2020 16:06, Paul Durrant wrote:
>> From: Paul Durrant <pdurrant@amazon.com>
>>
>> ... to specify a separate migration stream that will also be suitable for
>> live update.
>>
>> The original scope of the document was to support non-cooperative 
>> migration
>> of guests [1] but, since then, live update of xenstored has been 
>> brought into
>> scope. Thus it makes more sense to define a separate image format for
>> serializing xenstore state that is suitable for both purposes.
>>
>> The document has been limited to specifying a new image format. The 
>> mechanism
>> for acquiring the image for live update or migration is not covered as 
>> that
>> is more appropriately dealt with by a patch to docs/misc/xenstore.txt. 
>> It is
>> also expected that, when the first implementation of live update or 
>> migration
>> making use of this specification is committed, that the document is 
>> moved from
>> docs/designs into docs/specs.
>>
>> NOTE: It will only be necessary to save and restore state for active 
>> xenstore
>>        connections, but the documentation for 'RESUME' in xenstore.txt 
>> implies
>>        otherwise. That command is unused so this patch deletes it from 
>> the
>>        specification.

Could someone from Citrix please verify that XAPI isn't using XS_RESUME?


Juergen
Edwin Török May 5, 2020, 3:15 p.m. UTC | #3
On Tue, 2020-05-05 at 14:13 +0100, Jürgen Groß wrote:
> On 05.05.20 15:01, Julien Grall wrote:
> > Hi Paul,
> > 
> > On 28/04/2020 16:06, Paul Durrant wrote:
> > > From: Paul Durrant <pdurrant@amazon.com>
> > > 
> > > ... to specify a separate migration stream that will also be
> > > suitable for
> > > live update.
> > > 
> > > The original scope of the document was to support non-
> > > cooperative 
> > > migration
> > > of guests [1] but, since then, live update of xenstored has been 
> > > brought into
> > > scope. Thus it makes more sense to define a separate image format
> > > for
> > > serializing xenstore state that is suitable for both purposes.
> > > 
> > > The document has been limited to specifying a new image format.
> > > The 
> > > mechanism
> > > for acquiring the image for live update or migration is not
> > > covered as 
> > > that
> > > is more appropriately dealt with by a patch to
> > > docs/misc/xenstore.txt. 
> > > It is
> > > also expected that, when the first implementation of live update
> > > or 
> > > migration
> > > making use of this specification is committed, that the document
> > > is 
> > > moved from
> > > docs/designs into docs/specs.
> > > 
> > > NOTE: It will only be necessary to save and restore state for
> > > active 
> > > xenstore
> > >        connections, but the documentation for 'RESUME' in
> > > xenstore.txt 
> > > implies
> > >        otherwise. That command is unused so this patch deletes it
> > > from 
> > > the
> > >        specification.
> 
> Could someone from Citrix please verify that XAPI isn't using
> XS_RESUME?

The implementation of XS_RESUME in oxenstored doesn't do much: it seems
to be just a way for Dom0 to check whether a domain exists or not, and
for a domain to check whether they are Dom0 or not.
If the domain exists, then the resume implementation just returns `()`,
i.e. does nothing.

I can't find any references to Xs.resume in xenopsd (or the other XAPI
repos that I got checked out), so I think it can be safely removed from
the spec and client libraries (I'd keep it in the actual oxenstored
implementation just in case some guest does call it).

Best regards,
--Edwin

> 
> 
> Juergen
>
Jürgen Groß May 5, 2020, 3:26 p.m. UTC | #4
On 05.05.20 17:15, Edwin Torok wrote:
> On Tue, 2020-05-05 at 14:13 +0100, Jürgen Groß wrote:
>> On 05.05.20 15:01, Julien Grall wrote:
>>> Hi Paul,
>>>
>>> On 28/04/2020 16:06, Paul Durrant wrote:
>>>> From: Paul Durrant <pdurrant@amazon.com>
>>>>
>>>> ... to specify a separate migration stream that will also be
>>>> suitable for
>>>> live update.
>>>>
>>>> The original scope of the document was to support non-
>>>> cooperative
>>>> migration
>>>> of guests [1] but, since then, live update of xenstored has been
>>>> brought into
>>>> scope. Thus it makes more sense to define a separate image format
>>>> for
>>>> serializing xenstore state that is suitable for both purposes.
>>>>
>>>> The document has been limited to specifying a new image format.
>>>> The
>>>> mechanism
>>>> for acquiring the image for live update or migration is not
>>>> covered as
>>>> that
>>>> is more appropriately dealt with by a patch to
>>>> docs/misc/xenstore.txt.
>>>> It is
>>>> also expected that, when the first implementation of live update
>>>> or
>>>> migration
>>>> making use of this specification is committed, that the document
>>>> is
>>>> moved from
>>>> docs/designs into docs/specs.
>>>>
>>>> NOTE: It will only be necessary to save and restore state for
>>>> active
>>>> xenstore
>>>>         connections, but the documentation for 'RESUME' in
>>>> xenstore.txt
>>>> implies
>>>>         otherwise. That command is unused so this patch deletes it
>>>> from
>>>> the
>>>>         specification.
>>
>> Could someone from Citrix please verify that XAPI isn't using
>> XS_RESUME?
> 
> The implementation of XS_RESUME in oxenstored doesn't do much: it seems
> to be just a way for Dom0 to check whether a domain exists or not, and
> for a domain to check whether they are Dom0 or not.
> If the domain exists, then the resume implementation just returns `()`,
> i.e. does nothing.
> 
> I can't find any references to Xs.resume in xenopsd (or the other XAPI
> repos that I got checked out), so I think it can be safely removed from
> the spec and client libraries (I'd keep it in the actual oxenstored
> implementation just in case some guest does call it).

Thanks for the confirmation.


Juergen
diff mbox series

Patch

diff --git a/docs/designs/xenstore-migration.md b/docs/designs/xenstore-migration.md
index 6ab351e8fe..34a2afd17e 100644
--- a/docs/designs/xenstore-migration.md
+++ b/docs/designs/xenstore-migration.md
@@ -3,254 +3,399 @@ 
 ## Background
 
 The design for *Non-Cooperative Migration of Guests*[1] explains that extra
-save records are required in the migrations stream to allow a guest running
-PV drivers to be migrated without its co-operation. Moreover the save
-records must include details of registered xenstore watches as well as
-content; information that cannot currently be recovered from `xenstored`,
-and hence some extension to the xenstore protocol[2] will also be required.
-
-The *libxenlight Domain Image Format* specification[3] already defines a
-record type `EMULATOR_XENSTORE_DATA` but this is not suitable for
-transferring xenstore data pertaining to the domain directly as it is
-specified such that keys are relative to the path
-`/local/domain/$dm_domid/device-model/$domid`. Thus it is necessary to
-define at least one new save record type.
+save records are required in the migrations stream to allow a guest running PV
+drivers to be migrated without its co-operation. Moreover the save records must
+include details of registered xenstore watches as well as content; information
+that cannot currently be recovered from `xenstored`, and hence some extension
+to the xenstored implementations will also be required.
+
+As a similar set of data is needed for transferring xenstore data from one
+instance to another when live updating xenstored this document proposes an
+image format for a 'migration stream' suitable for both purposes.
 
 ## Proposal
 
-### New Save Record
+The image format consists of a _header_ followed by 1 or more _records_. Each
+record consists of a type and length field, followed by any data mandated by
+the record type. At minimum there will be a single record of type `END`
+(defined below).
 
-A new mandatory record type should be defined within the libxenlight Domain
-Image Format:
+### Header
 
-`0x00000007: DOMAIN_XENSTORE_DATA`
+The header identifies the stream as a `xenstore` stream, including the version
+of the specification that it complies with.
 
-An arbitrary number of these records may be present in the migration
-stream and may appear in any order. The format of each record should be as
-follows:
+All fields in this header must be in _big-endian_ byte order, regardless of
+the setting of the endianness bit.
 
 
 ```
     0       1       2       3       4       5       6       7    octet
 +-------+-------+-------+-------+-------+-------+-------+-------+
-| type                          | record specific data          |
-+-------------------------------+                               |
-...
-+---------------------------------------------------------------+
+| ident                                                         |
++-------------------------------+-------------------------------|
+| version                       | flags                         |
++-------------------------------+-------------------------------+
 ```
 
-where type is one of the following values
 
+| Field     | Description                                       |
+|-----------|---------------------------------------------------|
+| `ident`   | 0x78656e73746f7265 ('xenstore' in ASCII)          |
+|           |                                                   |
+| `version` | 0x00000001 (the version of the specification)     |
+|           |                                                   |
+| `flags`   | 0 (LSB): Endianness: 0 = little, 1 = big          |
+|           |                                                   |
+|           | 1-31: Reserved (must be zero)                     |
 
-| Field  | Description                                      |
-|--------|--------------------------------------------------|
-| `type` | 0x00000000: invalid                              |
-|        | 0x00000001: NODE_DATA                            |
-|        | 0x00000002: WATCH_DATA                           |
-|        | 0x00000003: TRANSACTION_DATA                     |
-|        | 0x00000004 - 0xFFFFFFFF: reserved for future use |
+### Records
 
+Records immediately follow the header and have the following format:
 
-and data is one of the record data formats described in the following
-sections.
+
+```
+    0       1       2       3       4       5       6       7    octet
++-------+-------+-------+-------+-------+-------+-------+-------+
+| type                          | len                           |
++-------------------------------+-------------------------------+
+| body
+...
+|       | padding (0 to 7 octets)                               |
++-------+-------------------------------------------------------+
+```
+
+NOTE: padding octets here and in all subsequent format specifications must be
+      written as zero and should be ignored when the stream is read.
 
 
-NOTE: The record data does not contain an overall length because the
-libxenlight record header specifies the length.
+| Field  | Description                                          |
+|--------|------------------------------------------------------|
+| `type` | 0x00000000: END                                      |
+|        | 0x00000001: GLOBAL_DATA                              |
+|        | 0x00000002: CONNECTION_DATA                          |
+|        | 0x00000003: WATCH_DATA                               |
+|        | 0x00000004: TRANSACTION_DATA                         |
+|        | 0x00000005: NODE_DATA                                |
+|        | 0x00000006 - 0xFFFFFFFF: reserved for future use     |
+|        |                                                      |
+| `len`  | The length (in octets) of `body`                     |
+|        |                                                      |
+| `body` | The type-specific record data                        |
 
+Some records will depend on other records in the migration stream. Records
+upon which other records depend must always appear earlier in the stream.
 
-**NODE_DATA**
+The various formats of the type-specific data are described in the following
+sections:
 
+\pagebreak
 
-Each NODE_DATA record specifies a single node in xenstore and is formatted
-as follows:
+### END
 
+The end record marks the end of the image, and is the final record
+in the stream.
 
 ```
-    0       1       2       3     octet
-+-------+-------+-------+-------+
-| NODE_DATA                     |
-+-------------------------------+
-| path length                   |
-+-------------------------------+
-| path data                     |
-...
-| pad (0 to 3 octets)           |
-+-------------------------------+
-| perm count (N)                |
-+-------------------------------+
-| perm0                         |
-+-------------------------------+
-...
-+-------------------------------+
-| permN                         |
-+-------------------------------+
-| value length                  |
-+-------------------------------+
-| value data                    |
-...
-| pad (0 to 3 octets)           |
-+-------------------------------+
+    0       1       2       3       4       5       6       7    octet
++-------+-------+-------+-------+-------+-------+-------+-------+
 ```
 
-where perm0..N are formatted as follows:
 
+The end record contains no fields; its body length is 0.
+
+\pagebreak
+
+### GLOBAL_DATA
+
+This record is only relevant for live update. It contains details of global
+xenstored state that needs to be restored.
 
 ```
-    0       1       2       3     octet
+    0       1       2       3    octet
 +-------+-------+-------+-------+
-| perm  | pad   | domid         |
+| rw-socket-fd                  |
++-------------------------------+
+| ro-socket-fd                  |
 +-------------------------------+
 ```
 
 
-path length and value length are specified in octets (excluding the NUL
-terminator of the path). perm should be one of the ASCII values `w`, `r`,
-`b` or `n` as described in [2]. All pad values should be 0.
-All paths should be absolute (i.e. start with `/`) and as described in
-[2].
+| Field          | Description                                  |
+|----------------|----------------------------------------------|
+| `rw-socket-fd` | The file descriptor of the socket accepting  |
+|                | read-write connections                       |
+|                |                                              |
+| `ro-socket-fd` | The file descriptor of the socket accepting  |
+|                | read-only connections                        |
+
+xenstored will resume in the original process context. Hence `rw-socket-fd` and
+`ro-socket-fd` simply specify the file descriptors of the sockets. Sockets
+are not always used, however, and so -1 will be used to denote an unused
+socket.
 
 
-**WATCH_DATA**
+\pagebreak
 
+### CONNECTION_DATA
 
-Each WATCH_DATA record specifies a registered watch and is formatted as
-follows:
+For live update the image format will contain a `CONNECTION_DATA` record for
+each connection to xenstore. For migration it will only contain a record for
+the domain being migrated.
 
 
 ```
-    0       1       2       3     octet
-+-------+-------+-------+-------+
-| WATCH_DATA                    |
-+-------------------------------+
-| wpath length                  |
-+-------------------------------+
-| wpath data                    |
-...
-| pad (0 to 3 octets)           |
-+-------------------------------+
+    0       1       2       3       4       5       6       7    octet
++-------+-------+-------+-------+-------+-------+-------+-------+
+| conn-id                       | conn-type     | conn-spec
 ...
++-------------------------------+-------------------------------+
+| data-len                      | data
 +-------------------------------+
-| token length                  |
-+-------------------------------+
-| token data                    |
 ...
-| pad (0 to 3 octets)           |
-+-------------------------------+
 ```
 
-wpath length and token length are specified in octets (excluding the NUL
-terminator). The wpath should be as described for the `WATCH` operation in
-[2]. The token is an arbitrary string of octets not containing any NUL
-values.
 
+| Field       | Description                                     |
+|-------------|-------------------------------------------------|
+| `conn-id`   | A non-zero number used to identify this         |
+|             | connection in subsequent connection-specific    |
+|             | records                                         |
+|             |                                                 |
+| `conn-type` | 0x0000: shared ring                             |
+|             | 0x0001: socket                                  |
+|             | 0x0002 - 0xFFFF: reserved for future use        |
+|             |                                                 |
+| `conn-spec` | See below                                       |
+|             |                                                 |
+| `data-len`  | The length (in octets) of any pending data not  |
+|             | yet written to the connection                   |
+|             |                                                 |
+| `data`      | Pending data (may be empty)                     |
 
-**TRANSACTION_DATA**
+The format of `conn-spec` is dependent upon `conn-type`.
 
+\pagebreak
 
-Each TRANSACTION_DATA record specifies an open transaction and is formatted
-as follows:
+For `shared ring` connections it is as follows:
 
 
 ```
-    0       1       2       3     octet
-+-------+-------+-------+-------+
-| TRANSACTION_DATA              |
-+-------------------------------+
-| tx_id                         |
-+-------------------------------+
+    0       1       2       3       4       5       6       7    octet
+                                                +-------+-------+
+                                                | flags         |
++---------------+---------------+---------------+---------------+
+| domid         | tdomid        | evtchn                        |
++-------------------------------+-------------------------------+
 ```
 
-where tx_id is the non-zero identifier values of an open transaction.
 
+| Field     | Description                                       |
+|-----------|---------------------------------------------------|
+| `domid`   | The domain-id that owns the shared page           |
+|           |                                                   |
+| `tdomid`  | The domain-id that `domid` acts on behalf of if   |
+|           | it has been subject to an SET_TARGET              |
+|           | operation [2] or DOMID_INVALID [3] otherwise      |
+|           |                                                   |
+| `flags`   | Must be zero                                      |
+|           |                                                   |
+| `evtchn`  | The port number of the interdomain channel used   |
+|           | by `domid` to communicate with xenstored          |
+|           |                                                   |
 
-### Protocol Extension
+Since the ABI guarantees that entry 1 in `domid`'s grant table will always
+contain the GFN of the shared page.
 
-Before xenstore state is migrated it is necessary to wait for any pending
-reads, writes, watch registrations etc. to complete, and also to make sure
-that xenstored does not start processing any new requests (so that new
-requests remain pending on the shared ring for subsequent processing on the
-new host). Hence the following operation is needed:
+For `socket` connections it is as follows:
 
-```
-QUIESCE                 <domid>|
 
-Complete processing of any request issued by the specified domain, and
-do not process any further requests from the shared ring.
+```
+                                                +-------+-------+
+                                                | flags         |
++---------------+---------------+---------------+---------------+
+| socket-fd                     | pad                           |
++-------------------------------+-------------------------------+
 ```
 
-The `WATCH` operation does not allow specification of a `<domid>`; it is
-assumed that the watch pertains to the domain that owns the shared ring
-over which the operation is passed. Hence, for the tool-stack to be able
-to register a watch on behalf of a domain a new operation is needed:
 
-```
-ADD_DOMAIN_WATCHES      <domid>|<watch>|+
+| Field       | Description                                     |
+|-------------|-------------------------------------------------|
+| `flags`     | A bit-wise OR of:                               |
+|             | 0001: read-only                                 |
+|             |                                                 |
+| `socket-fd` | The file descriptor of the connected socket     |
 
-Adds watches on behalf of the specified domain.
+This type of connection is only relevant for live update, where the xenstored
+resumes in the original process context. Hence `socket-fd` simply specify
+the file descriptor of the socket connection.
 
-<watch> is a NUL separated tuple of <path>|<token>. The semantics of this
-operation are identical to the domain issuing WATCH <path>|<token>| for
-each <watch>.
-```
+\pagebreak
+
+### WATCH_DATA
+
+The image format will contain a `WATCH_DATA` record for each watch registered
+by a connection for which there is `CONNECTION_DATA` record previously present.
 
-The watch information for a domain also needs to be extracted from the
-sending xenstored so the following operation is also needed:
 
 ```
-GET_DOMAIN_WATCHES      <domid>|<index>   <gencnt>|<watch>|*
+    0       1       2       3    octet
++-------+-------+-------+-------+
+| conn-id                       |
++---------------+---------------+
+| wpath-len     | token-len     |
++---------------+---------------+
+| wpath
+...
+| token
+...
+```
+
+
+| Field       | Description                                     |
+|-------------|-------------------------------------------------|
+| `conn-id`   | The connection that issued the `WATCH`          |
+|             | operation [2]                                   |
+|             |                                                 |
+| `wpath-len` | The length (in octets) of `wpath` including the |
+|             | NUL terminator                                  |
+|             |                                                 |
+| `token-len` | The length (in octets) of `token` including the |
+|             | NUL terminator                                  |
+|             |                                                 |
+| `wpath`     | The watch path, as specified in the `WATCH`     |
+|             | operation                                       |
+|             |                                                 |
+| `token`     | The watch identifier token, as specified in the |
+|             | `WATCH` operation                               |
+
+\pagebreak
 
-Gets the list of watches that are currently registered for the domain.
+### TRANSACTION_DATA
 
-<watch> is a NUL separated tuple of <path>|<token>. The sub-list returned
-will start at <index> items into the the overall list of watches and may
-be truncated (at a <watch> boundary) such that the returned data fits
-within XENSTORE_PAYLOAD_MAX.
+The image format will contain a `TRANSACTION_DATA` record for each transaction
+that is pending on a connection for which there is `CONNECTION_DATA` record
+previously present.
 
-If <index> is beyond the end of the overall list then the returned sub-
-list will be empty. If the value of <gencnt> changes then it indicates
-that the overall watch list has changed and thus it may be necessary
-to re-issue the operation for previous values of <index>.
+
+```
+    0       1       2       3    octet
++-------+-------+-------+-------+
+| conn-id                       |
++-------------------------------+
+| tx-id                         |
++-------------------------------+
 ```
 
-To deal with transactions that were pending when the domain is migrated
-it is necessary to start transactions with the same tx_id on behalf of the
-domain in the receiving xenstored.
 
-NOTE: For safety each such transaction should result in an `EAGAIN` when
-the `TRANSACTION_END` operation is performed, as modifications made under
-the tx_id will not be part of the migration stream.
+| Field          | Description                                  |
+|----------------|----------------------------------------------|
+| `conn-id`      | The connection that issued the               |
+|                | `TRANSACTION_START` operation [2]            |
+|                |                                              |
+| `tx-id`        | The transaction id passed back to the domain |
+|                | by the `TRANSACTION_START` operation         |
 
-The `TRANSACTION_START` operation does not allow specification of a
-`<domid>`; it is assumed that the transaction pertains to the domain that
-owns the shared ring over which the operation is passed. Neither does it
-allow a `<transid>` to be specified; it is always chosen by xenstored.
-Hence, for the tool-stack to be able to open a transaction on behalf of a
-domain a new operation is needed:
+\pagebreak
 
+### NODE_DATA
+
+For live update the image format will contain a `NODE_DATA` record for each
+node in xenstore. For migration it will only contain a record for the nodes
+relating to the domain being migrated. The `NODE_DATA` may be related to
+a _committed_ node (globally visible in xenstored) or a _pending_ node (created
+or modified by a transaction for which there is also a `TRANSACTION_DATA`
+record previously present).
+
+
+```
+    0       1       2       3    octet
++-------+-------+-------+-------+
+| conn-id                       |
++-------------------------------+
+| tx-id                         |
++---------------+---------------+
+| path-len      | value-len     |
++---------------+---------------+
+| access        | perm-count    |
++---------------+---------------+
+| perm1                         |
++-------------------------------+
+...
++-------------------------------+
+| permN                         |
++---------------+---------------+
+| path
+...
+| value
+...
 ```
-START_DOMAIN_TRANSACTION    <domid>|<transid>|
 
-Starts a transaction on behalf of a domain.
 
-The semantics of this are similar to the domain issuing
-TRANSACTION_START and receiving the specified <transid> as the response.
-The main difference is that the transaction will be immediately marked as
-'conflicting' such that when the domain issues TRANSACTION_END T|, it will
-result in EAGAIN.
+| Field        | Description                                    |
+|--------------|------------------------------------------------|
+| `conn-id`    | If this value is non-zero then this record     |
+|              | related to a pending transaction               |
+|              |                                                |
+| `tx-id`      | This value should be ignored if `conn-id` is   |
+|              | zero. Otherwise it specifies the id of the     |
+|              | pending transaction                            |
+|              |                                                |
+| `path-len`   | The length (in octets) of `path` including the |
+|              | NUL terminator                                 |
+|              |                                                |
+| `value-len`  | The length (in octets) of `value` (which will  |
+|              | be zero for a deleted node)                    |
+|              |                                                |
+| `access`     | This value should be ignored if this record    |
+|              | does not relate to a pending transaction,      |
+|              | otherwise it specifies the accesses made to    |
+|              | the node and hence is a bitwise OR of:         |
+|              |                                                |
+|              | 0x0001: read                                   |
+|              | 0x0002: written                                |
+|              |                                                |
+|              | The value will be zero for a deleted node      |
+|              |                                                |
+| `perm-count` | The number (N) of node permission specifiers   |
+|              | (which will be 0 for a node deleted in a       |
+|              | pending transaction)                           |
+|              |                                                |
+| `perm1..N`   | A list of zero or more node permission         |
+|              | specifiers (see below)                         |
+|              |                                                |
+| `path`       | The absolute path of the node                  |
+|              |                                                |
+| `value`      | The node value (which may be empty or contain  |
+|              | NUL octets)                                    |
+
+
+A node permission specifier has the following format:
+
+
+```
+    0       1       2       3    octet
++-------+-------+-------+-------+
+| perm  | pad   | domid         |
++-------+-------+---------------+
 ```
 
-It may also be desirable to state in the protocol specification that
-the `INTRODUCE` operation should not clear the `<gfn>` specified such that
-a `RELEASE` operation followed by an `INTRODUCE` operation form an
-idempotent pair. The current implementation of *C xentored* does this
-(in the `domain_conn_reset()` function) but this could be dropped as this
-behaviour is not currently specified and the page will always be zeroed
-for a newly created domain.
+| Field   | Description                                         |
+|---------|-----------------------------------------------------|
+| `perm`  | One of the ASCII values `w`, `r`, `b` or `n` as     |
+|         | specified for the `SET_PERMS` operation [2]         |
+|         |                                                     |
+| `domid` | The domain-id to which the permission relates       |
 
+Note that perm1 defines the domain owning the code. See [4] for more
+explanation of node permissions.
 
 * * *
 
 [1] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/designs/non-cooperative-migration.md
+
 [2] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/misc/xenstore.txt
-[3] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/specs/libxl-migration-stream.pandoc
+
+[3] See https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=xen/include/public/xen.h;hb=HEAD#l612
+
+[4] https://wiki.xen.org/wiki/XenBus
\ No newline at end of file
diff --git a/docs/misc/xenstore.txt b/docs/misc/xenstore.txt
index 04ce0ba607..cb8009cb68 100644
--- a/docs/misc/xenstore.txt
+++ b/docs/misc/xenstore.txt
@@ -289,23 +289,6 @@  IS_DOMAIN_INTRODUCED	<domid>|		T| or F|
 	ie, if INTRODUCE for the domain has not yet been followed by
 	domain destruction or explicit RELEASE.
 
-RESUME			<domid>|
-
-	Arranges that @releaseDomain events will once more be
-	generated when the domain becomes shut down.  This might have
-	to be used if a domain were to be shut down (generating one
-	@releaseDomain) and then subsequently restarted, since the
-	state-sensitive algorithm in xenstored will not otherwise send
-	further watch event notifications if the domain were to be
-	shut down again.
-
-	It is not clear whether this is possible since one would
-	normally expect a domain not to be restarted after being shut
-	down without being destroyed in the meantime.  There are
-	currently no users of this request in xen-unstable.
-
-	xenstored prevents the use of RESUME other than by dom0.
-
 SET_TARGET		<domid>|<tdomid>|
 	Notifies xenstored that domain <domid> is targeting domain
 	<tdomid>. This grants domain <domid> full access to paths